**Estimation of Energy Storage and Its Feasibility Analysis**

Mohammad Taufiqul Arif, Amanullah M. T. Oo and A. B. M. Shawkat Ali

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/52218

## **1. Introduction**

40 Energy Storage – Technologies and Applications

l'énergie; Avril 2003.

mars 2005.

[91] Multon B, Ruer J. Stocker l'électricité : Oui, c'est indispensable, et c'est possible ! pourquoi, où, comment?. Publication ECRIN en contribution au débat national sur

[92] Robyns B. Contribution du stockage de l'énergie électrique à la participation au services système des éoliennes. Séminaire SRBE – SEE – L2EP « Éolien et réseaux : enjeux », 22

[93] Multon B. L'énergie électrique : analyse des ressources et de la production. Journées

section électrotechnique du club EEA. Paris, France, 28-29 janvier 1999, p. 8.

Storage significantly adds flexibility in Renewable Energy (RE) and improves energy management. This chapter explains the estimation procedures of required storage with grid connected RE to support for a residential load. It was considered that storage integrated RE will support all the steady state load and grid will support transient high loads. This will maximize the use of RE. Proper sized RE resources with proper sized storage is essential for best utilization of RE in a cost effective way. This chapter also explains the feasibility analysis of storage by comparing the economical and environmental indexes.

Most of the presently installed Solar PV or Wind turbines are without storage while connected to the grid. The intermittent nature of solar radiation and wind speed limits the capacity of RE to follow the load demand. The available standards described sizing and requirements of storage in standalone systems. However standards available for distributed energy resources (DER) or distributed resources (DR) to connect to the grid while considering solar photovoltaic (PV), wind turbine and storage as DR. Bearing this limitation, this chapter followed the sizing guidelines for standalone system to estimate the required storage for the grid connected RE applications.

Solar PV is unable provide electricity during night and cloudy days; similarly wind energy also unable to follow load demand. Moreover PV and/or wind application is not able to follow the load demand; when these RE generators are just in the stage to start generating energy and when these RE are in highest mode of generating stage while load demand falls to the lowest level. Therefore it can be said that RE is unable to generate energy by following the load demand which is a major limitation in energy management. Storage can play this critical role of proper energy management. Moreover storage helps in reducing the intermittent nature of RE and improve the Power Quality (PQ). This study considers regional Australia as the study area also considered residential load, solar radiation and

© 2013 Arif et al., licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2013 Arif et al., licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

wind speed data of that location for detailed analysis. Figure 1 shows the daily load profile (summer: January 01, 2009 and winter: July 01 2009) of Capricornia region of Rockhampton, a regional city in Australia. Ergon Energy [1] is the utility operator in that area. However load demand of the residential load in that area depends on the work time patter which is different than the load profile of Figure 1. Overall electricity demand is very high in the evening and also in the morning for the residential load however PV generates electricity mostly during the day time therefore residents need to purchase costly electricity during peak demand in the evening. Similarly wind energy also unable to follow the residential load profile. Therefore properly estimated storage needs to be integrated to overcome this situation.

Estimation of Energy Storage and Its Feasibility Analysis 43

unplanned way that even with battery integrated system is not able to support the load in reliable way. Figure 2 illustrates a typical situation when whole system in jeopardize as the

estimation of storage system was not done correctly.

**Figure 2.** Typical condition of failure in storage integrated RE system

The adoption of storage with the PV system certainly incurs additional cost to the system but the benefits of adding storage has not been clearly assessed. Therefore this chapter aims to achieve two objectives. One is to estimate the required storage for the grid connected PV system or grid connected wind turbine or combination of grid connected PV and wind turbine system to achieve the maximum daily use of RE. Second objective is to identify the effects of storage on the designed system in terms of environment and economic by comparing the same system with and without storage. The feasibility of the designed system is expressed as, the Cost of Energy (COE) is closer to the present system while providing environmental benefits by reducing Greenhouse Gas (GHG) emission and improving the Renewable Fraction (RF).

Data was collected for the Capricornia region of Rockhampton city in Queensland, Australia. Load data was collected for a 3 bed room house by estimating all the electrical appliances demand and average usage period considering its ratings. Daily load profile drawn from hourly load data and total daily load was estimated by calculating the area under the daily load profile curve using trapezoidal method. Weather data was collected for the year 2009 from [2] for this location and calculated the energy output from PV array and

**Figure 1.** Daily load profile of Capricornia region in 2009 (Summer & Winter)

This chapter explores the need of storage systems to maximize the use of RE, furthermore estimates the required capacity of storage to meet the daily need which will gradually eliminate the dependency on conventional energy sources. Estimation of storage sizing is explained in section 3. This chapter also conducts the feasibility assessment of storage in terms of economic and environmental perspective which is explained in section 4.

## **2. Background**

Solar and Wind are the two major sources of RE. Australia is one of best places for these sources. In regional areas of Australia, roof top Solar PV is installed in many residential houses either in off-grid or grid connected configurations and most residential wind turbine are for specific applications in off-grid configuration. In grid connected solar PV systems where storage is not integrated, the energy output from this system does not satisfy to the desired level. Currently installed most of the residential PV systems are designed in an unplanned way that even with battery integrated system is not able to support the load in reliable way. Figure 2 illustrates a typical situation when whole system in jeopardize as the estimation of storage system was not done correctly.

**Figure 2.** Typical condition of failure in storage integrated RE system

42 Energy Storage – Technologies and Applications

situation.

**2. Background** 

wind speed data of that location for detailed analysis. Figure 1 shows the daily load profile (summer: January 01, 2009 and winter: July 01 2009) of Capricornia region of Rockhampton, a regional city in Australia. Ergon Energy [1] is the utility operator in that area. However load demand of the residential load in that area depends on the work time patter which is different than the load profile of Figure 1. Overall electricity demand is very high in the evening and also in the morning for the residential load however PV generates electricity mostly during the day time therefore residents need to purchase costly electricity during peak demand in the evening. Similarly wind energy also unable to follow the residential load profile. Therefore properly estimated storage needs to be integrated to overcome this

**Figure 1.** Daily load profile of Capricornia region in 2009 (Summer & Winter)

This chapter explores the need of storage systems to maximize the use of RE, furthermore estimates the required capacity of storage to meet the daily need which will gradually eliminate the dependency on conventional energy sources. Estimation of storage sizing is explained in section 3. This chapter also conducts the feasibility assessment of storage in

Solar and Wind are the two major sources of RE. Australia is one of best places for these sources. In regional areas of Australia, roof top Solar PV is installed in many residential houses either in off-grid or grid connected configurations and most residential wind turbine are for specific applications in off-grid configuration. In grid connected solar PV systems where storage is not integrated, the energy output from this system does not satisfy to the desired level. Currently installed most of the residential PV systems are designed in an

terms of economic and environmental perspective which is explained in section 4.

The adoption of storage with the PV system certainly incurs additional cost to the system but the benefits of adding storage has not been clearly assessed. Therefore this chapter aims to achieve two objectives. One is to estimate the required storage for the grid connected PV system or grid connected wind turbine or combination of grid connected PV and wind turbine system to achieve the maximum daily use of RE. Second objective is to identify the effects of storage on the designed system in terms of environment and economic by comparing the same system with and without storage. The feasibility of the designed system is expressed as, the Cost of Energy (COE) is closer to the present system while providing environmental benefits by reducing Greenhouse Gas (GHG) emission and improving the Renewable Fraction (RF).

Data was collected for the Capricornia region of Rockhampton city in Queensland, Australia. Load data was collected for a 3 bed room house by estimating all the electrical appliances demand and average usage period considering its ratings. Daily load profile drawn from hourly load data and total daily load was estimated by calculating the area under the daily load profile curve using trapezoidal method. Weather data was collected for the year 2009 from [2] for this location and calculated the energy output from PV array and wind turbine. Figure 3 shows the solar radiation and wind speed of Rockhampton for the year 2009. Hourly energy output curve of PV and wind was drawn and compared with the load profile to estimate the required storage. For estimation, choice was taken considering the worst month weather data and it was observed that May to July is worst period when solar and wind have lower energy density as shown in Figure 3.

**Figure 3.** Solar radiation and wind speed in Rockhampton in 2009

#### **Solar power**

PV array considered as a device that produces DC electricity in direct proportion to the global solar radiation. Therefore, the power output of the PV array can be calculated by Equation 1[3-4].

$$\mathbf{P\_{PV}} = \mathbf{Y\_{PV}}\mathbf{f\_{PV}} \left(\frac{\mathbf{\bar{G}\_T}}{\mathbf{\bar{G}\_{T,STC}}}\right) \left[1 + \alpha\_\mathbf{P} \left(\mathbf{T\_C} - \mathbf{T\_{C,STC}}\right)\right] \tag{1}$$

If there is no effect of temperature on the PV array, the temperature coefficient of the power is zero, thus the above equation can be simplified as Equation 2 [3-4].

$$\mathbf{P\_{PV}} = \mathbf{Y\_{PV}} \mathbf{f\_{PV}}(\frac{\mathbf{\bar{G}\_T}}{\mathbf{\bar{G}\_{T,STC}}}) \tag{2}$$

Estimation of Energy Storage and Its Feasibility Analysis 45

� ρAV� (3)

)� sq.m (4)

)∝ (5)

) (6)

C�ρAV� (7)

P = �

in m/s.

length = D/2) can be calculated by Equation 4.

vertical wind profile Equation 5 [6-7]:

expressed in Europe by Equation 6 [7]:

0.59. Therefore Equation-3 can be written as:

**2.1. Estimation of daily residential load** 

windbreaks is 0.03m.

(4.85m/s) [8].

where, *P* is Power output from wind turbine in Watts, *ρ* is the air density (1.225kg/m3 at 15οC and 1-atmosphere or in sea level), *A* is rotor swept area in m2 and *V* is the wind speed

The swept area of a horizontal axis wind turbine of rotor diameter (D) in meter (or blade

As power in the wind is proportional to the cube of the wind speed therefore increase in wind speed is very significant. One way to get more power is by increasing the tower height. Hourly wind speed at different height above ground level can be calculated by the

V� = V�(

V� = V�(

where v1 and v2 are the wind speeds at heights H1 and H2 and α is the wind shear component or power law exponent or friction coefficient. A typical value of α is 0.14 for countryside or flat plane area. Equation 5 commonly used in United States and the same is

where z is the roughness length in meters. A typical value of z for open area with a few

Temperature has effect on air density which changes the output of wind turbine. Average wind speed globally at 80m height is higher during day time (4.96m/s) than night time

However German physicist Albert Betz concluded in 1919 that no wind turbine can convert more than 16/27 or 59.3% of the kinetic energy of the wind into mechanical energy by turning a rotor. This is the maximum theoretical efficiency of rotor and this is known as Betz Limit or Betz' Law. This is also called "power coefficient" and the maximum value is: *CP*=

The following subsections describe the residential load, solar & wind energy of

Preferred method of determining load is bottom-up approach in which daily load is anticipated and summed to yield an average daily load. This can be done by multiplying the

P = � �

Rockhampton area and also describe the importance of storage.

�� ��

��(��⁄�) ��(��⁄�)

A = π(� �

where *YPV* - rated capacity of PV array, meaning power output under standard test conditions [kW]; *fPV* - PV de-rating factor [%]; *GT* - solar radiation incident on PV array in current time step [kW/m2]; *GT,STC* - incident radiation under standard test conditions [1 kW/m2]; *α<sup>P</sup>* - temperature coefficient of power [%/οC]; *TC* - PV cell temperature in current time step [οC]; *TC,STC* -PV cell temperature under standard test conditions [25οC]. Performance of PV array depends on derating factors like temperature, dirt and mismatched modules.

#### **Wind Power**

Kinetic energy of wind can be converted into electrical energy by using wind turbine, rotor, gear box and generator. The available power of wind is the flux of kinetic energy, which the air is interacting with rotor per unit time at a cross sectional area of the rotor and that can be expressed [5] as per Equation 3:

Estimation of Energy Storage and Its Feasibility Analysis 45

$$P = \frac{1}{2}\rho A V^3 \tag{3}$$

where, *P* is Power output from wind turbine in Watts, *ρ* is the air density (1.225kg/m3 at 15οC and 1-atmosphere or in sea level), *A* is rotor swept area in m2 and *V* is the wind speed in m/s.

The swept area of a horizontal axis wind turbine of rotor diameter (D) in meter (or blade length = D/2) can be calculated by Equation 4.

$$\mathbf{A} = \pi (\frac{\mathbf{b}}{2})^2 \text{ sq.m.} \tag{4}$$

As power in the wind is proportional to the cube of the wind speed therefore increase in wind speed is very significant. One way to get more power is by increasing the tower height. Hourly wind speed at different height above ground level can be calculated by the vertical wind profile Equation 5 [6-7]:

$$\mathbf{V\_2} = \mathbf{V\_1(}^{\mathbf{H\_2}}\_{\mathbf{H\_1}})^{\text{oc}} \tag{5}$$

where v1 and v2 are the wind speeds at heights H1 and H2 and α is the wind shear component or power law exponent or friction coefficient. A typical value of α is 0.14 for countryside or flat plane area. Equation 5 commonly used in United States and the same is expressed in Europe by Equation 6 [7]:

$$\mathbf{V\_2} = \mathbf{V\_1(}^{\ln(\mathbf{H\_2/z})}{\ln(\mathbf{H\_1/z})}\text{\AA}\tag{6}$$

where z is the roughness length in meters. A typical value of z for open area with a few windbreaks is 0.03m.

Temperature has effect on air density which changes the output of wind turbine. Average wind speed globally at 80m height is higher during day time (4.96m/s) than night time (4.85m/s) [8].

However German physicist Albert Betz concluded in 1919 that no wind turbine can convert more than 16/27 or 59.3% of the kinetic energy of the wind into mechanical energy by turning a rotor. This is the maximum theoretical efficiency of rotor and this is known as Betz Limit or Betz' Law. This is also called "power coefficient" and the maximum value is: *CP*= 0.59. Therefore Equation-3 can be written as:

$$P = \frac{1}{2} \mathcal{C}\_{\text{P}} \rho \text{AV}^3 \tag{7}$$

The following subsections describe the residential load, solar & wind energy of Rockhampton area and also describe the importance of storage.

#### **2.1. Estimation of daily residential load**

44 Energy Storage – Technologies and Applications

wind turbine. Figure 3 shows the solar radiation and wind speed of Rockhampton for the year 2009. Hourly energy output curve of PV and wind was drawn and compared with the load profile to estimate the required storage. For estimation, choice was taken considering the worst month weather data and it was observed that May to July is worst period when

PV array considered as a device that produces DC electricity in direct proportion to the global solar radiation. Therefore, the power output of the PV array can be calculated by

If there is no effect of temperature on the PV array, the temperature coefficient of the power

P�� = Y��f��( ���

where *YPV* - rated capacity of PV array, meaning power output under standard test conditions [kW]; *fPV* - PV de-rating factor [%]; *GT* - solar radiation incident on PV array in current time step [kW/m2]; *GT,STC* - incident radiation under standard test conditions [1 kW/m2]; *α<sup>P</sup>* - temperature coefficient of power [%/οC]; *TC* - PV cell temperature in current time step [οC]; *TC,STC* -PV cell temperature under standard test conditions [25οC]. Performance of PV array depends on

Kinetic energy of wind can be converted into electrical energy by using wind turbine, rotor, gear box and generator. The available power of wind is the flux of kinetic energy, which the air is interacting with rotor per unit time at a cross sectional area of the rotor and that can be

�������

� �� � ����T� − T������� (1)

) (2)

�������

solar and wind have lower energy density as shown in Figure 3.

**Figure 3.** Solar radiation and wind speed in Rockhampton in 2009

P�� = Y��f�� � ���

is zero, thus the above equation can be simplified as Equation 2 [3-4].

derating factors like temperature, dirt and mismatched modules.

**Solar power** 

Equation 1[3-4].

**Wind Power** 

expressed [5] as per Equation 3:

Preferred method of determining load is bottom-up approach in which daily load is anticipated and summed to yield an average daily load. This can be done by multiplying the

power rating of all the appliances by the number of hours it is expected to operate on an average day to obtain Watt-hour (Wh) value as shown in Table 1. The load data collected from a 3 bed room house in Kawana, Rockhampton in Australia and total land area of the house is 700m2 where 210m2 is the building area with available roof space. For grid connected household appliances daily average load can also be obtained from monthly utility bills.

Estimation of Energy Storage and Its Feasibility Analysis 47

(8)

weekday from Monday to Friday. A 24 hour load profile of a particular day as shown in Figure 4. It was found that maximum load demand was in the evening from 6:00PM to

Hourly load is a time series data and total daily load can be estimated by calculating the

Daily Load = P���� <sup>=</sup> � f(x)dt ���

Where pt1 = Load (in kW) at time t1 =1 in hour, pt2 = Load (in kW) at time t2 = 2 in hour, T12 =

Following Equation 8 total daily AC (Alternating current) load is the area under this load curve which is 15.7kWh and the equivalent DC (Direct current) load is shown considering

Solar radiation varies with time and season. For estimation of available useful solar energy, worst month solar radiation was considered to ensure that the designed system can operate year-round. In Australia yearly average sunlight hours varies from 5 to 10 hours/day and maximum area is over 8 hours/day [2]. From the collected data it was found that in Rockhampton solar radiation over 5.0kWh/m2/d varies from 08:00AM to 16:00PM i.e. sun

The daily average solar radiation of Kawana suburb in the Capricornia region of Rockhampton city is as shown in the Figure 5. It was found that annual average solar radiation was 5.48kWh/m2/day. Lowest monthly average solar radiation was 4kWh/m2/day

��

10:00PM and in the morning 7:00AM to 9:00AM.

**Figure 4.** Daily load profile of a Residential house

� (p�� + p��)T��

time difference b/w t1 and t2 in hour.

Where f(x) <sup>=</sup> �

hour is 8hrs/day.

area under the load profile curve using Equation 8.

efficiency of the converter as 85% which is 18.47kWh.

**2.2. Estimation of daily available solar energy** 


**Table 1.** Daily load consumption of a house

*Data source: Product catalogue and [7]*

Load profile of a residential house varies according to the residents work time pattern. Working nature of the residents of Kawana suburb is such that most of the residents start for work between 7:00AM to 8:00AM and returns home between 5:00PM to 6:00PM during weekday from Monday to Friday. A 24 hour load profile of a particular day as shown in Figure 4. It was found that maximum load demand was in the evening from 6:00PM to 10:00PM and in the morning 7:00AM to 9:00AM.

**Figure 4.** Daily load profile of a Residential house

Hourly load is a time series data and total daily load can be estimated by calculating the area under the load profile curve using Equation 8.

$$\text{Daily Load} = \text{ }P\_{\text{Load}} = \int\_{\text{t}\_1}^{\text{t}\_2 \text{s}} \text{f}(\text{x})d\text{t} \tag{8}$$

Where f(x) <sup>=</sup> � � (p�� + p��)T��

46 Energy Storage – Technologies and Applications

Refrigerator 602kWh/year

(300W)

utility bills.

Washing machine (vertical axis)

Air conditioner (Window)

TV 32" LCD (Active/Standby)

DVD player (Active/Standby)

Computer (Laptop)

power rating of all the appliances by the number of hours it is expected to operate on an average day to obtain Watt-hour (Wh) value as shown in Table 1. The load data collected from a 3 bed room house in Kawana, Rockhampton in Australia and total land area of the house is 700m2 where 210m2 is the building area with available roof space. For grid connected household appliances daily average load can also be obtained from monthly

**Appliances Rating Daily time of use Qty Daily use** 

500W Weekends (1hr/week) 1 71

1200W Summer night & Holidays (1hr) 3 1200

150/3.5W Morning & night (4 hrs) 1 670

17/5.9W Night (2 hrs) 1 50

20W Night (4 - 5hrs) 1 80

Freezer 88W Whole day 1 880 Electrical Stove 2100W Morning & Evening (1-2hrs) 1 2100 Microwave Oven 1000W Morning & Evening (30 min to 1 hr) 1 500 Rice cooker 400W Evening (30 minutes) 1 200 Toaster 800W Morning (10 - 30 minutes) 1 80 Ceiling Fan 65W Summer night (4 -5 hrs) & Holidays 5 1300 Fluorescent light 16W Night (6 - 8 hours) 20 320

Vacuum Cleaner 1400W Weekends (1hr/week) 1 200

Cordless phone 4W Whole day 1 96

Clothe iron 1400W Night & Holidays (15 - 30 minutes) 1 350

Hot Water System 1800W Whole day( 3- 4 hrs) 1 5400 **Total: 15,747** 

Load profile of a residential house varies according to the residents work time pattern. Working nature of the residents of Kawana suburb is such that most of the residents start for work between 7:00AM to 8:00AM and returns home between 5:00PM to 6:00PM during

minutes)

Heater (Portable) 1200W Winter night & Holidays (30

**Table 1.** Daily load consumption of a house

*Data source: Product catalogue and [7]*

Whole day 1 1650

**(Wh/day)** 

1 600

Where pt1 = Load (in kW) at time t1 =1 in hour, pt2 = Load (in kW) at time t2 = 2 in hour, T12 = time difference b/w t1 and t2 in hour.

Following Equation 8 total daily AC (Alternating current) load is the area under this load curve which is 15.7kWh and the equivalent DC (Direct current) load is shown considering efficiency of the converter as 85% which is 18.47kWh.

#### **2.2. Estimation of daily available solar energy**

Solar radiation varies with time and season. For estimation of available useful solar energy, worst month solar radiation was considered to ensure that the designed system can operate year-round. In Australia yearly average sunlight hours varies from 5 to 10 hours/day and maximum area is over 8 hours/day [2]. From the collected data it was found that in Rockhampton solar radiation over 5.0kWh/m2/d varies from 08:00AM to 16:00PM i.e. sun hour is 8hrs/day.

The daily average solar radiation of Kawana suburb in the Capricornia region of Rockhampton city is as shown in the Figure 5. It was found that annual average solar radiation was 5.48kWh/m2/day. Lowest monthly average solar radiation was 4kWh/m2/day on May and highest solar radiation was from October to December in 2009. PV system designed to supply entire load considering the worst month solar radiation, which will deliver sufficient energy during rest of the year.

Estimation of Energy Storage and Its Feasibility Analysis 49

**Month Daily Time period Time window (hrs)**

Jan 06:00 - 20:00 14 Feb 03:00 - 17:00 14 Mar 00:00 - 15:00, 22:00 - 24:00 17 Apr 00:00 - 04:00, 20:00 - 24:00 8 May 16:00 - 24:00 8 Jun 12:00 - 19:00 7 Jul 10:00 - 15:00 5 Aug 07:00 - 12:00 5 Sep 01:00 - 10:00 9 Oct 00:00 - 06:00, 19:00 - 24:00 11 Nov 00:00 - 03:00, 15:00 - 24:00 12 Dec 13:00 - 24:00 11

**Table 2.** Wind speed period or window (6m/s or more)

**Figure 6.** Wind speed at 10m, 40m and 80m height in Rockhampton

**Figure 7.** Energy converted per m2 area at 40m height wind speed in Rockhampton

For estimating daily solar energy, worst month (May) solar radiation was considered and Figure 5 shows hourly solar radiation for May 07, 2009. Daily total solar energy was estimated by calculating area under the solar radiation curve using Equation 8. Therefore total solar radiation in May 07, 2009 was 1.582975kWh/m2/d. This energy generated by 1m2 PV area. Total solar radiation will increase with the increased area of the PV array.

**Figure 5.** Daily solar radiation (May 07, 2009) in Rockhampton

## **2.3. Estimation of daily available wind energy**

Wind speed varies with different natural factors, time and season. To estimate the available useful wind energy, worst month wind speed was considered to ensure that the designed system can operate year-round. From the collected data of Rockhampton it was found that July had the worst wind speed scenario as shown in Figure 3. It was found that in 2009, wind speed of Rockhampton was 6m/s or more for daily average duration of 10 hours. However for the month of July and August it was only 5 hours as shown in Table 2. Therefore wind speed data of July was considered for estimation of daily energy.

Three hourly wind speed data at 10.4m above sea level was collected from [2] for the year 2009, which was interpolated to get hourly data. At rotor height of 10m, 40m and 80m corresponding wind speed as shown in Figure 6. For energy estimation, July 03, 2009 wind speed data was considered and corresponding energy was calculated for 1m2 of rotor wind area at 40m rotor height using Equation 7 as shown in Figure 7. Betz limit, gearbox, bearing and generator efficiency was considered and overall efficiency of the wind turbine was taken 25%. Total energy output from wind turbine on July 03, 2009 is the area under the curve of Figure 7 (11:00AM to 09:00PM) which is 0.232785kWh/m2/d.


**Table 2.** Wind speed period or window (6m/s or more)

deliver sufficient energy during rest of the year.

**Figure 5.** Daily solar radiation (May 07, 2009) in Rockhampton

**2.3. Estimation of daily available wind energy** 

on May and highest solar radiation was from October to December in 2009. PV system designed to supply entire load considering the worst month solar radiation, which will

For estimating daily solar energy, worst month (May) solar radiation was considered and Figure 5 shows hourly solar radiation for May 07, 2009. Daily total solar energy was estimated by calculating area under the solar radiation curve using Equation 8. Therefore total solar radiation in May 07, 2009 was 1.582975kWh/m2/d. This energy generated by 1m2

Wind speed varies with different natural factors, time and season. To estimate the available useful wind energy, worst month wind speed was considered to ensure that the designed system can operate year-round. From the collected data of Rockhampton it was found that July had the worst wind speed scenario as shown in Figure 3. It was found that in 2009, wind speed of Rockhampton was 6m/s or more for daily average duration of 10 hours. However for the month of July and August it was only 5 hours as shown in Table 2.

Three hourly wind speed data at 10.4m above sea level was collected from [2] for the year 2009, which was interpolated to get hourly data. At rotor height of 10m, 40m and 80m corresponding wind speed as shown in Figure 6. For energy estimation, July 03, 2009 wind speed data was considered and corresponding energy was calculated for 1m2 of rotor wind area at 40m rotor height using Equation 7 as shown in Figure 7. Betz limit, gearbox, bearing and generator efficiency was considered and overall efficiency of the wind turbine was taken 25%. Total energy output from wind turbine on July 03, 2009 is the area under the

Therefore wind speed data of July was considered for estimation of daily energy.

curve of Figure 7 (11:00AM to 09:00PM) which is 0.232785kWh/m2/d.

PV area. Total solar radiation will increase with the increased area of the PV array.

**Figure 6.** Wind speed at 10m, 40m and 80m height in Rockhampton

**Figure 7.** Energy converted per m2 area at 40m height wind speed in Rockhampton

## **2.4. Importance of storage**

The ability to store large amounts of energy would allow electrical utilities to have greater flexibility in their operation, because with this option the supply demand do not have to be matched instantaneously [9]. Table 3 shows the benefits and limitations of different storage systems.

Estimation of Energy Storage and Its Feasibility Analysis 51

A recent study on high penetration of PV on present grid, mentioned that energy storage is the ultimate solution for allowing intermittent sources to address utility base load needs [15]. Storage integrated PV/Wind systems provides a combination of operational, financial

Improper sized PV/Wind system is unable to meet the load requirements, sometimes electrical energy from RE wasted which neither can be used by the load nor can be stored in battery. This event occurs when the battery State of Charge (SOC) exceeds its maximum allowable value and the solar/wind power output exceeds load demand. The amount of wasted/lost energy can be avoided or reduced by proper choice of battery and PV/Wind generation sizes. G.B. Shrestha et.al. in [16] mentioned that PV panel size and the battery size have different impacts on the indices of performance and proper balance between the two is necessary. A proper match between the installed capacities with the load demand is

Brahmi Nabiha et. al. in [17] presents sizing of mini autonomous hybrid grid, including PV, wind, generator and battery. The performance of any battery, expressed essentially by the voltage, load capacity and SOC or the Depth of Discharge (DOD). The usable energy in a

IEEE Std-1013-2007 [18] provides the recommendations for sizing of lead-acid batteries for stand-alone PV systems. This recommended practice provides a systematic approach for determining the appropriate energy capacity of a lead-acid battery to satisfy the energy requirements of the load for residential, commercial and industrial stand-alone PV systems. IEEE Std-1561-2007 [19]provides guideline for optimizing the performance and life of Lead-Acid batteries in remote hybrid power systems; which includes PV, wind, batteries. It also explains the battery sizing considerations for the application. IEEE Std 1547-2003 [20] provides guideline to connect Distributed Resources (DR), such as PV, wind and storage with the power grid at the distribution level. Grid connected system sizing for storage

Considering the above sizing practices and guidelines Figure 8 shows the steps for estimation of required storage for steady state residential load. For the easy of this analysis both PV and Wind turbine are considered to produce DC power which than converted to

E������ = C x V��� x DOD��� (9)

and environmental benefits.

**3. Estimation of storage sizing** 

essential to optimize such installation.

battery can be expressed by Equation 9.

integrated PV system also explained in [7].

AC by inverter, also considered battery as storage device.

**Step 1.** Determine the daily load of a residential house

**Step 2.** Determine the required PV or Wind turbine rating for the load **Step 3.** Determine daily energy output from the PV array or Wind turbine

The following steps are summarized for estimation.

where *C* is battery capacity and *Vbat* is the battery cell voltage.


**Table 3.** advantage and limitations of few storage systems [12-14] a. charge energy factor b. Fuel heat rate

The role of Energy Storage (ES) with Renewable Electricity generation is mentioned in[10] that the selection of ES system depends on application which is largely determined by the length of discharge. Based on the length of discharge, ES applications are often divided into three categories named power quality, bridging power and energy management applications. Although large scale storage is still expensive but research is going on for inexpensive and efficient batteries [11] suitable for large scale RE applications.

RE can be considered for different kinds of applications i.e. from small stand-alone remote systems to large scale grid-connected solar/wind power application. However development goes on to remote areas and brings the remote areas close to the grid network and eventually connected to the power grid and these RE generator are expected to operate as grid connected Distributed Generator (DG). Grid connected PV/wind with battery as storage can provide future-proof energy autonomy and allow home or office to generate clean energy and supply extra energy to the grid.

A recent study on high penetration of PV on present grid, mentioned that energy storage is the ultimate solution for allowing intermittent sources to address utility base load needs [15]. Storage integrated PV/Wind systems provides a combination of operational, financial and environmental benefits.

## **3. Estimation of storage sizing**

50 Energy Storage – Technologies and Applications

The ability to store large amounts of energy would allow electrical utilities to have greater flexibility in their operation, because with this option the supply demand do not have to be matched instantaneously [9]. Table 3 shows the benefits and limitations of different storage

**Storage Advantage Limitations** 

*Response time (S)* 

minutes

90 30 0.01 Low energy density.

25 360 Location specific.

Expensive to build.

Expensive to build.

Large standby loss

Expensive

Expensive.

Expensive.

Storage tank is expensive

*Life time (years)* 

Flywheel 85 20 0.1 Low energy density.

75 30 Tens of

Batteries 80 10 0.01 Early stage technology.

Capacitor 80 10 0.01 Low energy density.

Hydrogen 50 25 360 Highly flammable

inexpensive and efficient batteries [11] suitable for large scale RE applications.

The role of Energy Storage (ES) with Renewable Electricity generation is mentioned in[10] that the selection of ES system depends on application which is largely determined by the length of discharge. Based on the length of discharge, ES applications are often divided into three categories named power quality, bridging power and energy management applications. Although large scale storage is still expensive but research is going on for

RE can be considered for different kinds of applications i.e. from small stand-alone remote systems to large scale grid-connected solar/wind power application. However development goes on to remote areas and brings the remote areas close to the grid network and eventually connected to the power grid and these RE generator are expected to operate as grid connected Distributed Generator (DG). Grid connected PV/wind with battery as storage can provide future-proof energy autonomy and allow home or office to generate clean

Pumped hydro 80 50 10 Location specific.

*Efficiency (%)* 

*a fcef=1.3 bffhr=4300kJ/k*

*Wh* 

**Table 3.** advantage and limitations of few storage systems [12-14]

a. charge energy factor b. Fuel heat rate

energy and supply extra energy to the grid.

**2.4. Importance of storage** 

Compressed air energy

storage (CAES)

Thermal energy storage(TES)

Superconducting magnetic energy storage (SMES)

systems.

Improper sized PV/Wind system is unable to meet the load requirements, sometimes electrical energy from RE wasted which neither can be used by the load nor can be stored in battery. This event occurs when the battery State of Charge (SOC) exceeds its maximum allowable value and the solar/wind power output exceeds load demand. The amount of wasted/lost energy can be avoided or reduced by proper choice of battery and PV/Wind generation sizes. G.B. Shrestha et.al. in [16] mentioned that PV panel size and the battery size have different impacts on the indices of performance and proper balance between the two is necessary. A proper match between the installed capacities with the load demand is essential to optimize such installation.

Brahmi Nabiha et. al. in [17] presents sizing of mini autonomous hybrid grid, including PV, wind, generator and battery. The performance of any battery, expressed essentially by the voltage, load capacity and SOC or the Depth of Discharge (DOD). The usable energy in a battery can be expressed by Equation 9.

$$\mathbf{E}\_{\text{usable}} = \mathbf{C} \ge \mathbf{V}\_{\text{bat}} \ge \text{DOD}\_{\text{max}} \tag{9}$$

where *C* is battery capacity and *Vbat* is the battery cell voltage.

IEEE Std-1013-2007 [18] provides the recommendations for sizing of lead-acid batteries for stand-alone PV systems. This recommended practice provides a systematic approach for determining the appropriate energy capacity of a lead-acid battery to satisfy the energy requirements of the load for residential, commercial and industrial stand-alone PV systems. IEEE Std-1561-2007 [19]provides guideline for optimizing the performance and life of Lead-Acid batteries in remote hybrid power systems; which includes PV, wind, batteries. It also explains the battery sizing considerations for the application. IEEE Std 1547-2003 [20] provides guideline to connect Distributed Resources (DR), such as PV, wind and storage with the power grid at the distribution level. Grid connected system sizing for storage integrated PV system also explained in [7].

Considering the above sizing practices and guidelines Figure 8 shows the steps for estimation of required storage for steady state residential load. For the easy of this analysis both PV and Wind turbine are considered to produce DC power which than converted to AC by inverter, also considered battery as storage device.

The following steps are summarized for estimation.


**Step 4.** Estimate PV array size and wind turbine rotor diameter

**Step 5.** Compare the daily energy output (from PV or wind turbine) with the daily load, find the required load that storage needs to support

Estimation of Energy Storage and Its Feasibility Analysis 53

<sup>8</sup> = 1.9625kW

**3.1. Estimation of storage for grid connected residential solar PV** 

P��(kW) <sup>=</sup> Energy (kWh day) <sup>⁄</sup>

η = inverter efficiency \* dirty collector \* mismatched modules = 85%

So the adjusted PV array size for the equivalent DC load becomes:

storage required to support its load for 24 hours a day.

A =

area.

P������ <sup>=</sup>P��(kW)

Equivalent DC load on PV array can be found by considering the efficiency as:

The size of the PV array is determined by the daily average load divided by the available solar window or sun-hours per day. Generally, grid connected PV systems are designed to provide from 10% to 60% of energy needs with the difference being supplied from power utility[21]. However PV contribution can be increased to 100% of average steady state load.

Daily extractable solar energy as calculated in section 2.2 was considered for the estimation of required PV size for the residential load. Daily load of a three bed room house as calculated in section 2.1 is 15.7kWh. Therefore the PV array should support at least 15.7kWh of load everyday at the solar energy rate of 1.582975kWh/m2/day. Solar window is 8 hours or more in Rockhampton [2], therefore the required PV array capacity for the AC load as:

Solar window (h day) <sup>⁄</sup> <sup>=</sup> 15.7

<sup>η</sup> <sup>=</sup> 1.9625

To use battery as storage system, size of the PV array needs to be more than 1.3 times the load [18] in stand-alone configuration. But for the grid connected configuration 1.0 or 1.1 is good enough to avoid over design. For this designed residential load, it was considered 1.1.

P������(��������) = 1.1xP������ = 1.1x2.31 = 2.541kW Therefore, for this three bed room house 2.541kW capacity of PV array with proper sized

For known PV efficiency and for 1kW/m2 rated PV module required surface area of the PV array can be calculated. The efficiency of crystal silicon PV module is 12.5% [7], however LG

P������ = (1kW m� ⁄ )insolation ∗ A ∗ η

Therefore 20.328m2 PV module with PV array efficiency of 12.5% will support the load with sufficient storage size. This PV area is much smaller than the area of the designed house roof

This PV size was considered to calculate the total energy from PV array and to estimate the required storage for the load. Batteries last longer if they are shallow cycled. The capacity of

1x0.125 = 20.328m�

Polycrystalline PV module efficiency is 13.7% [22], therefore surface area becomes:

(1kW m⁄ �)η <sup>=</sup> 2.541

P������

0.85 = 2.31kW

Following the steps in Figure 8, the estimation starts by calculating required PV size.

**Step 6.** For the load on storage estimate the required Battery/Storage size in Ah.

The following sub-sections describe the estimation of required storage for grid connected PV, Wind and hybrid systems considering the residential load of Rockhampton as estimated in section 2.1.

**Figure 8.** Storage size estimation steps

#### **3.1. Estimation of storage for grid connected residential solar PV**

52 Energy Storage – Technologies and Applications

Collect wind speed Data

Find daily effective wind hour (h/day) for the site

Calculate daily energy in 1m2 wind area (kWh/m2/d)

**Figure 8.** Storage size estimation steps

in section 2.1.

Wind data Collection &

Wind turbine size or Rotor diameter

Storage size determination

determination

Load calculation

**Step 4.** Estimate PV array size and wind turbine rotor diameter

find the required load that storage needs to support

**Step 5.** Compare the daily energy output (from PV or wind turbine) with the daily load,

The following sub-sections describe the estimation of required storage for grid connected PV, Wind and hybrid systems considering the residential load of Rockhampton as estimated

Calculate the AC load on PV/Wind gen (kW)

)( *REWindow h day Dailyload kWh day Pac kW*

Convert AC load to DC load (at STC) Pdc,STC = Pac(kW) /Inverter efficiency

Pdc,STC(Adjusted) = Pdc,STC x n (according to IEEE-1013, n>1.3)

Calculate PV array size in require surface area Area of PV array (A) = Pdc,STC(Adjusted) /(1kW/m2 x η) or Calculate required rotor diameter for rated wind turbine capacity by calculating required swept area

Calculate total energy from Solar radiation for the PV area Total Solar radiation = Solar radiation (kWh/m<sup>2</sup>

2 /( ) <sup>3</sup> *CPA <sup>p</sup> pV*

or Calculate total energy output from estimated wind turbine

2/)( <sup>3</sup> *CP <sup>p</sup> pAV*

By plotting daily load curve and total daily energy output from solar/wind generator calculate the load on Storage Load on storage = (Area under the curve of total solar/wind output curve – Common area of load and solar/wind curve)

Set the time/days or autonomy period for storage support Set the nominal DC system voltage of the Storage/Battery Calculate the load in Ah on Storage per day: Total load (Ah/day @ system voltage) = Total DC load (Wh/d)/System Voltage (V) Calculate the required Storage size for the load Usable Storage (Ah) = Total load (Ah/d) x Required time (day)

Considering discharge rate with temperature and MDOD, calculate total required Storage capacity by: Total Storage capacity (Ah) = Usable Storage (Ah) / (MDOD x (T, DR)

Collect or Calculate Daily load (kWh) Collect Solar radiation Data

Find daily effective

solar window (h/

day) for the site

) x A

)/( )/( Calculate energy from daily solar radiation (kWh/m2

/d)

Solar data Collection &

Load calculation

PV array size determination

Storage size determination

**Step 6.** For the load on storage estimate the required Battery/Storage size in Ah.

The size of the PV array is determined by the daily average load divided by the available solar window or sun-hours per day. Generally, grid connected PV systems are designed to provide from 10% to 60% of energy needs with the difference being supplied from power utility[21]. However PV contribution can be increased to 100% of average steady state load. Following the steps in Figure 8, the estimation starts by calculating required PV size.

Daily extractable solar energy as calculated in section 2.2 was considered for the estimation of required PV size for the residential load. Daily load of a three bed room house as calculated in section 2.1 is 15.7kWh. Therefore the PV array should support at least 15.7kWh of load everyday at the solar energy rate of 1.582975kWh/m2/day. Solar window is 8 hours or more in Rockhampton [2], therefore the required PV array capacity for the AC load as:

$$P\_{\rm ac} \text{(kW)} = \frac{\text{Energy (kWh/day)}}{\text{Solar window (h/day)}} = \frac{15.7}{8} = 1.9625 \text{kW}$$

Equivalent DC load on PV array can be found by considering the efficiency as:

η = inverter efficiency \* dirty collector \* mismatched modules = 85%

$$P\_{\rm dc,STC} = \frac{P\_{\rm ac} \,\mathrm{(kW)}}{\eta} = \frac{1.9625}{0.85} = 2.31 \mathrm{kW}$$

To use battery as storage system, size of the PV array needs to be more than 1.3 times the load [18] in stand-alone configuration. But for the grid connected configuration 1.0 or 1.1 is good enough to avoid over design. For this designed residential load, it was considered 1.1. So the adjusted PV array size for the equivalent DC load becomes:

$$\mathbf{P\_{dc,STC(Adjusted)}} = \mathbf{1.1xP\_{dc,STC}} = \mathbf{1.1x2.31} = \mathbf{2.541kW}$$

Therefore, for this three bed room house 2.541kW capacity of PV array with proper sized storage required to support its load for 24 hours a day.

For known PV efficiency and for 1kW/m2 rated PV module required surface area of the PV array can be calculated. The efficiency of crystal silicon PV module is 12.5% [7], however LG Polycrystalline PV module efficiency is 13.7% [22], therefore surface area becomes:

$$\text{P}\_{\text{dc,STC}} = (1 \text{kW}/\text{m}^2) \text{insolation} \ast \text{A} \ast \eta$$

$$\text{A} = \frac{\text{P}\_{\text{dc,STC}}}{(1 \text{kW}/\text{m}^2)\eta} = \frac{2.541}{1 \times 0.125} = 20.328 \text{m}^2$$

Therefore 20.328m2 PV module with PV array efficiency of 12.5% will support the load with sufficient storage size. This PV area is much smaller than the area of the designed house roof area.

This PV size was considered to calculate the total energy from PV array and to estimate the required storage for the load. Batteries last longer if they are shallow cycled. The capacity of

the battery bank can be calculated by multiplying the daily load on battery by the autonomy day or the number of days it should provide power continuously. The ampere-hour (Ah) rating of the battery bank can be found after dividing the battery bank capacity by the battery bank voltage (e.g. 24V or 48V). It is generally not recommended to design for more than 12 days of autonomy for off-grid system and for grid connected system one day autonomy is good to design.

Estimation of Energy Storage and Its Feasibility Analysis 55

Calendar Life (Year)

Efficiencies Ah% Wh%

3

Cycle Life (Cycles)

Daily maximum load in Ah @ system voltage =

Load (Wh/day) 25.98272x10 = = = 1082.613Ah/d System Voltage 24

Energy storage in a battery typically given in Ah, at system voltage and at some specified

Lead-acid, SLI 20% 50 500 1-2 90 75 Lead-acid, golf cart 80% 45 1000 3-5 90 75 Lead-acid, deep-cycle 80% 35 2000 7-10 90 75 Nickel-cadmium 100% 20 1000-2000 10-15 70 60 Nickel-metal hydride 100% 50 1000-2000 8-10 70 65

The Ah capacity of a battery is not only rate-dependent but also depends on temperature. The capacity under varying temperature and discharge rates to a reference condition of C/20 at 25οC is explained in [7]. Lead-acid battery capacity decreases dramatically in colder temperature conditions. However heat is also not good for batteries. In Rockhampton average temperature is above 20οC. The Maximum depth of discharge (MDOD) for Lead-

Load (Ah/day) x No of days 511.416x1 Battery storage (minimum) = = = 639.27Ah

Load (Ah/day) x No of days 1082.613x1 Battery storage (maximum) = <sup>=</sup> = 1353.26Ah

The rated capacity of battery is specified at standard temperature. At 25οC, the discharge rate of C/20 type battery (i.e. discharge for 20 hours), becomes 96% [7], therefore finally

Required minimum Battery storage(25 C,20hour-rate)=

*Ah*

Required maximum Battery storage(25 C,20hour-rate)=

*Ah*

Battery storage 639.27 <sup>=</sup> 665.90

Battery storage 1353.26 <sup>=</sup> 1409.64

Rated capacity 0.96

Rated capacity 0.96

MDOD 0.80

MDOD 0.80

°

°

discharge rate. Table 4 shows characteristics of several types of batteries.

(Wh/kg)

acid batteries is 80%, therefore for one day discharge the batteries need to store:

Battery type MDOD Energy Density

**Table 4.** Comparison of Battery Characteristics[7]

required battery capacity becomes:

Total solar energy generated by the 20.328m2 PV array at the solar radiation rate of Rockhampton is plotted in Figure 9 and calculated as 32.17872kWh which is the area under the PV output curve. Now superimpose the DC load curve on the PV output curve to find the load that needs to be supported by the storage as shown in Figure 9. The common area under the curve is 6.196kWh which is the area of the load that served by the PV array during day time while charging the batteries as well. The remaining load is (18.47 - 6.196) =12.274kWh/day that needs to be served by the storage. This is the daily minimum load on storage. However the design was based on to support total load, therefore the remaining energy from the PV array should be managed by the storage system which is (32.17872 - 6.196) = 25.98272kWh/day. This is the maximum load on storage, if total energy generated by PV array needs to be managed by the storage.

**Figure 9.** PV output and daily load curve shows the load on storage

Inverters are specified by their DC input voltage as well as by their AC output voltage, continuous power handling capability and the amount of surge power they can supply for brief periods of time. Inverter's DC input voltage which is the same as the voltage of the Battery bank and the PV array is called the system voltage. The system voltage usually considered as 12V, 24V or 48V. The system voltage for this designed DC system was considered 24V and this system was designed for one day. Considering inverter efficiency of 95% [23], the required battery capacity can be calculated.

Daily minimum load in Ah @ system voltage =

$$= \frac{\text{Load (Wh/day)}}{\text{System Voltage}} = \frac{12.274 \times 10^3}{24} = 511.416 \text{Ah/ds}$$

Daily maximum load in Ah @ system voltage =

$$=\frac{\text{Load (Wh/day)}}{\text{System Voltage}}=\frac{25.98272 \times 10^3}{24} = 1082.613 \text{Ah/ds}$$

Energy storage in a battery typically given in Ah, at system voltage and at some specified discharge rate. Table 4 shows characteristics of several types of batteries.


**Table 4.** Comparison of Battery Characteristics[7]

54 Energy Storage – Technologies and Applications

autonomy is good to design.

by PV array needs to be managed by the storage.

**Figure 9.** PV output and daily load curve shows the load on storage

95% [23], the required battery capacity can be calculated.

the battery bank can be calculated by multiplying the daily load on battery by the autonomy day or the number of days it should provide power continuously. The ampere-hour (Ah) rating of the battery bank can be found after dividing the battery bank capacity by the battery bank voltage (e.g. 24V or 48V). It is generally not recommended to design for more than 12 days of autonomy for off-grid system and for grid connected system one day

Total solar energy generated by the 20.328m2 PV array at the solar radiation rate of Rockhampton is plotted in Figure 9 and calculated as 32.17872kWh which is the area under the PV output curve. Now superimpose the DC load curve on the PV output curve to find the load that needs to be supported by the storage as shown in Figure 9. The common area under the curve is 6.196kWh which is the area of the load that served by the PV array during day time while charging the batteries as well. The remaining load is (18.47 - 6.196) =12.274kWh/day that needs to be served by the storage. This is the daily minimum load on storage. However the design was based on to support total load, therefore the remaining energy from the PV array should be managed by the storage system which is (32.17872 - 6.196) = 25.98272kWh/day. This is the maximum load on storage, if total energy generated

Inverters are specified by their DC input voltage as well as by their AC output voltage, continuous power handling capability and the amount of surge power they can supply for brief periods of time. Inverter's DC input voltage which is the same as the voltage of the Battery bank and the PV array is called the system voltage. The system voltage usually considered as 12V, 24V or 48V. The system voltage for this designed DC system was considered 24V and this system was designed for one day. Considering inverter efficiency of

Daily minimum load in Ah @ system voltage =

Load (Wh/day) 12.274x10 = = = 511.416Ah/d System Voltage 24

3

The Ah capacity of a battery is not only rate-dependent but also depends on temperature. The capacity under varying temperature and discharge rates to a reference condition of C/20 at 25οC is explained in [7]. Lead-acid battery capacity decreases dramatically in colder temperature conditions. However heat is also not good for batteries. In Rockhampton average temperature is above 20οC. The Maximum depth of discharge (MDOD) for Leadacid batteries is 80%, therefore for one day discharge the batteries need to store:

$$\text{Battery storage (minimum)} = \frac{\text{Load (Ah/day)} \times \text{No of days}}{\text{MDOD}} = \frac{511.416 \times 1}{0.80} = 639.27 \text{Ah}$$

$$\text{Battery storage (maximum)} = \frac{\text{Load (Ah/day)} \times \text{No of days}}{\text{MDOD}} = \frac{1082.613 \times 1}{0.80} = 1353.26 \text{Ah}$$

MDOD 0.80

The rated capacity of battery is specified at standard temperature. At 25οC, the discharge rate of C/20 type battery (i.e. discharge for 20 hours), becomes 96% [7], therefore finally required battery capacity becomes:

> Required minimum Battery storage(25 C,20hour-rate)= Battery storage 639.27 <sup>=</sup> 665.90 Rated capacity 0.96 ° *Ah* Required maximum Battery storage(25 C,20hour-rate)= Battery storage 1353.26 <sup>=</sup> 1409.64 Rated capacity 0.96 ° *Ah*

## **3.2. Estimation of storage for grid connected residential wind power**

Following the similar steps in section 3.1, required wind turbine capacity was calculated and then required storage was estimated for the same load of 15.7kWh/day.

Estimation of Energy Storage and Its Feasibility Analysis 57

**Figure 10.** Wind turbine output and daily load curve shows the load on storage

Battery storage (minimum) =

Battery storage (maximum) =

Battery storage 559.0625 <sup>=</sup> 582.356

Rated capacity 0.96

discharge the battery needs to store the energy as:

capacity becomes:

3

3

Daily minimum load in Ah @ system voltage =

Load (Wh/day) 10.734x10 = = = 447.25Ah/d System Voltage 24

Daily maximum load in Ah @ system voltage =

Load (Wh/day) 18.619x10 = = = 775.79Ah/d System Voltage 24

Energy storage in a battery typically given in Ah, at system voltage and at some specified discharge rate. Consider MDOD for Lead-Acid batteries is 80%, therefore for one day

> Load (Ah/day) x No of days 447.25x1 <sup>=</sup> = = 559.0625Ah MDOD 0.80

> Load (Ah/day) x No of days 775.79x1 <sup>=</sup> = = 969.7375Ah MDOD 0.80

The rated capacity of battery is specified at standard temperature. At 25οC, the discharge rate of C/20 (i.e. discharge for 20 hours), becomes 96% [7], therefore finally required battery

Required minimum Battery storage(25 C,20hour-rate)=

*Ah*

°

Energy generated by wind turbine at 40m height for 1m2 rotor wind area was calculated in section 2.3, which is 0.232785kWh/m2/d. The output of the wind turbine needs to be improved such that at least 15.7kWh of load should be supported each day. It was found that in July, wind speed was 6m/s or above only for 5hrs/day at 10m height, however at 40m height wind speed was 6m/s or above for 10hrs/day, therefore the rotor height was considered 40m. The required wind turbine size for the load can be calculated as:

$$P\_{\rm ac} \text{(kW)} = \frac{\text{Load (kWh/day)}}{\text{Window} \text{(h/day)}} = \frac{15.7}{10} = 1.57 \text{kW}$$

This estimation is for required storage which is a DC component; it requires inverter to support the load. DC capacity of the wind turbine can be calculated considering inverter efficiency of 90%.

$$P\_{\rm dc,STC} = \frac{P\_{\rm ac} \text{(kW)}}{\eta} = \frac{1.57}{0.90} = 1.744 \text{kW}$$

Likewise PV assumption, wind turbine capacity is considered 1.1 times the required load in grid connected configuration, to charge batteries while supporting load.

$$\mathbf{P\_{dc,STC(Adjusted)}} = \mathbf{1.1xP\_{dc,STC}} = \mathbf{1.1x1.744} = \mathbf{1.92kW}$$

Energy generated by wind turbine on July 03, 2009 was 0.232785kWh/m2/d. To support total load, rotor swept area needs to be adjusted. Equation 7 shows that power output is not linear for increase in rotor diameter. It was found that at 40m rotor height, wind speed varied b/w 6.17m/s to 9.92m/s, therefore average wind speed of 8m/s was considered to calculate the rotor diameter for the rated wind turbine capacity of 1.92kW. The rotor diameter was calculated as 5.58m and calculated total energy is 26.355kWh which is the area under the wind turbine output curve as shown in Figure 10.

Daily load curve was plotted on the daily energy output curve and calculated the common area to estimate the required load on storage to support for the day. It was found that 7.736kWh of load was supported by the wind turbine while charging the storage. The remaining (18.47 - 7.736) = 10.734kWh of load needs to be supported by the storage each day. This is the minimum load on storage. However the design was considering to manage 100% load therefore remaining (26.355 - 7.736) = 18.619kWh of energy must be managed by the storage. This is the maximum load on storage.

Considering the DC system voltage as 24V, load on battery in Ah can be calculated for one day as:

**Figure 10.** Wind turbine output and daily load curve shows the load on storage

efficiency of 90%.

day as:

**3.2. Estimation of storage for grid connected residential wind power** 

considered 40m. The required wind turbine size for the load can be calculated as:

then required storage was estimated for the same load of 15.7kWh/day.

P��(kW) <sup>=</sup> Load (kWh day) <sup>⁄</sup>

P������ <sup>=</sup>P��(kW)

grid connected configuration, to charge batteries while supporting load.

under the wind turbine output curve as shown in Figure 10.

the storage. This is the maximum load on storage.

Following the similar steps in section 3.1, required wind turbine capacity was calculated and

Energy generated by wind turbine at 40m height for 1m2 rotor wind area was calculated in section 2.3, which is 0.232785kWh/m2/d. The output of the wind turbine needs to be improved such that at least 15.7kWh of load should be supported each day. It was found that in July, wind speed was 6m/s or above only for 5hrs/day at 10m height, however at 40m height wind speed was 6m/s or above for 10hrs/day, therefore the rotor height was

Windwindow (h day) <sup>⁄</sup> <sup>=</sup> 15.7

This estimation is for required storage which is a DC component; it requires inverter to support the load. DC capacity of the wind turbine can be calculated considering inverter

<sup>η</sup> <sup>=</sup> 1.57

Likewise PV assumption, wind turbine capacity is considered 1.1 times the required load in

P������(��������) = 1.1xP������ = 1.1x1.744 = 1.92kW Energy generated by wind turbine on July 03, 2009 was 0.232785kWh/m2/d. To support total load, rotor swept area needs to be adjusted. Equation 7 shows that power output is not linear for increase in rotor diameter. It was found that at 40m rotor height, wind speed varied b/w 6.17m/s to 9.92m/s, therefore average wind speed of 8m/s was considered to calculate the rotor diameter for the rated wind turbine capacity of 1.92kW. The rotor diameter was calculated as 5.58m and calculated total energy is 26.355kWh which is the area

Daily load curve was plotted on the daily energy output curve and calculated the common area to estimate the required load on storage to support for the day. It was found that 7.736kWh of load was supported by the wind turbine while charging the storage. The remaining (18.47 - 7.736) = 10.734kWh of load needs to be supported by the storage each day. This is the minimum load on storage. However the design was considering to manage 100% load therefore remaining (26.355 - 7.736) = 18.619kWh of energy must be managed by

Considering the DC system voltage as 24V, load on battery in Ah can be calculated for one

0.90 = 1.744kW

<sup>10</sup> = 1.57kW

$$\text{Daily minimum load in Ah} \text{ @ system voltage} = \frac{\text{Load (Wh/day)}}{\text{System Voltage}} = \frac{10.734 \text{x} \text{10}^3}{24} = 447.25 \text{Ah/d}$$

Daily maximum load in Ah @ system voltage =

$$= \frac{\text{Load (Wh/day)}}{\text{System Voltage}} = \frac{18.619 \times 10^3}{24} = 775.79 \text{ Ah/dl}$$

Energy storage in a battery typically given in Ah, at system voltage and at some specified discharge rate. Consider MDOD for Lead-Acid batteries is 80%, therefore for one day discharge the battery needs to store the energy as:

$$\begin{aligned} & \text{Battery storage (minimum)} = \\ &= \frac{\text{Load (Ah/day)} \times \text{No of days}}{\text{MDOD}} = \frac{447.25 \times 1}{0.80} = 559.0625 \text{Ah} \\ &= \frac{\text{Battery storage (maximum)} }{\text{MDOD}} = \frac{775.79 \times 1}{0.80} = 969.7375 \text{Ah} \end{aligned}$$

The rated capacity of battery is specified at standard temperature. At 25οC, the discharge rate of C/20 (i.e. discharge for 20 hours), becomes 96% [7], therefore finally required battery capacity becomes:

$$\begin{aligned} &\text{Required minimum Battery storage} (25 \, ^\circ \text{C,20hour-rate}) = \\ &= \frac{\text{Battery storage}}{\text{Rated capacity}} = \frac{559.0625}{0.96} = 582.356 \, Ah \end{aligned}$$

Required maximum Battery storage(25 C,20hour-rate)= Battery storage 969.7375 <sup>=</sup> 1010.143 Rated capacity 0.96 ° *Ah*

Estimation of Energy Storage and Its Feasibility Analysis 59

<sup>10</sup> = 0.785kW

Therefore 10.16m2 of PV area required for this hybrid system. The output energy from this PV module is plotted in Figure 11. For the remaining load the required wind turbine is

For Wind turbine: 50% AC Load is (15.7/2) = 7.85kWh/d

Windwindow (h day) <sup>⁄</sup> <sup>=</sup> 7.85

The inverter considered with this wind turbine of efficiency 90%, therefore the DC capacity

<sup>η</sup> <sup>=</sup> 0.785

For this designed house load, wind turbine capacity considered 1.1 times the load. So the

P������(��������) = 1.1xP������ = 1.1x0.872 = 0.9592kW Average wind speed of 8m/s was considered to calculate the rotor diameter for the required capacity of wind turbine. For the 0.9592kW capacity wind turbine, the rotor diameter becomes 3.95m and daily energy generated by this wind turbine was plotted in Figure 11.

Total energy generated from this hybrid system is 28.12kWh and compared with the DC load it was calculated that the Hybrid system support directly 8.45kWh of load as shown the common area in Figure 11. Therefore the minimum (18.47 - 8.45) = 10.02kWh of load needs

0.90 = 0.872kW

P��(kW) <sup>=</sup> Load (kWh day) <sup>⁄</sup>

P������ <sup>=</sup>P��(kW)

**Figure 11.** Hybrid system output and daily load curve shows the load on storage

adjusted wind turbine size for the equivalent DC load becomes:

estimated as:

becomes:

Required wind turbine capacity becomes:

#### **3.3. Estimation of storage for grid connected residential hybrid system**

Many studies indicated that hybrid system is always better than any single RE system. However the practical implementation depends on the availability of adequate solar radiation, wind speed and their seasonal variation. Other critical point is adequate space for hybrid system installation and moreover the overall cost of the installation. The study location of this analysis is suitable for both solar and wind energy. It was found that for little variation of wind speed, convertible energy variation is much higher therefore wind energy fluctuation is higher than solar energy. Considering all the scenarios and for the easy of analysis it was considered that 50% of load to be supported by solar and 50% by wind energy.

Following the steps in Figure 8 and earlier sections, required storage is estimated.

For Solar PV: 50% AC Load is (15.7/2) = 7.85kWh/d

Required PV array capacity becomes:

$$P\_{\rm ac} \text{(kW)} = \frac{\text{Energy (kWh/day)}}{\text{Solar window (h/day)}} = \frac{7.85}{8} = 0.98125 \text{kW}$$

Equivalent DC load can be found by considering the efficiency of the PV system as:

η = inverter efficiency \* dirty collector \* mismatched modules = 85%

$$P\_{\rm dc,STC} = \frac{P\_{\rm ac} \text{(kW)}}{\eta} = \frac{0.98125}{0.85} = 1.1544 \text{kW}$$

For this designed house load, PV capacity considered 1.1 times the load. So the adjusted PV array size for the equivalent DC load becomes:

$$\mathbf{P\_{dc,STC(Adjusted)}} = \mathbf{1.1xP\_{dc,STC}} = \mathbf{1.1x1.1544} = \mathbf{1.27kW}$$

Therefore it requires 1.27kW capacity of PV array with proper sized storage to support 50% load for 24 hours a day.

Considering the crystal silicon PV module whose efficiency is 12.5% [7], therefore the surface area of PV module becomes:

$$P\_{\rm dc,STC} = (1 \text{kW/m}^2) \text{insolation} \ast \text{A} \ast \text{\eta}$$

$$\text{A} = \frac{\text{P}\_{\text{dc,STC}}}{(1 \text{kW}/\text{m}^2)\eta} = \frac{1.27}{1 \text{x} 0.125} = 10.16 \text{m}^2$$

Therefore 10.16m2 of PV area required for this hybrid system. The output energy from this PV module is plotted in Figure 11. For the remaining load the required wind turbine is estimated as:

$$\text{For Wind turbine: } 50\% \text{ AC Load is (15.7/2) = 7.85kWh/d.}$$

Required wind turbine capacity becomes:

58 Energy Storage – Technologies and Applications

Required PV array capacity becomes:

array size for the equivalent DC load becomes:

A =

P������

(1kW m⁄ �)η <sup>=</sup> 1.27

load for 24 hours a day.

surface area of PV module becomes:

energy.

Required maximum Battery storage(25 C,20hour-rate)=

Many studies indicated that hybrid system is always better than any single RE system. However the practical implementation depends on the availability of adequate solar radiation, wind speed and their seasonal variation. Other critical point is adequate space for hybrid system installation and moreover the overall cost of the installation. The study location of this analysis is suitable for both solar and wind energy. It was found that for little variation of wind speed, convertible energy variation is much higher therefore wind energy fluctuation is higher than solar energy. Considering all the scenarios and for the easy of analysis it was considered that 50% of load to be supported by solar and 50% by wind

*Ah*

°

<sup>8</sup> = 0.98125kW

0.85 = 1.1544kW

1x0.125 = 10.16m�

Battery storage 969.7375 <sup>=</sup> 1010.143

**3.3. Estimation of storage for grid connected residential hybrid system** 

Following the steps in Figure 8 and earlier sections, required storage is estimated.

P��(kW) <sup>=</sup> Energy (kWh day) <sup>⁄</sup>

η = inverter efficiency \* dirty collector \* mismatched modules = 85%

P������ <sup>=</sup>P��(kW)

For Solar PV: 50% AC Load is (15.7/2) = 7.85kWh/d

Solar window (h day) <sup>⁄</sup> <sup>=</sup> 7.85

<sup>η</sup> <sup>=</sup> 0.98125

For this designed house load, PV capacity considered 1.1 times the load. So the adjusted PV

P������(��������) = 1.1xP������ = 1.1x1.1544 = 1.27kW Therefore it requires 1.27kW capacity of PV array with proper sized storage to support 50%

Considering the crystal silicon PV module whose efficiency is 12.5% [7], therefore the

P������ = (1kW m� ⁄ )insolation ∗ A ∗ η

Equivalent DC load can be found by considering the efficiency of the PV system as:

Rated capacity 0.96

$$P\_{\rm ac} \text{(kW)} = \frac{\text{Load (kWh/day)}}{\text{Window} \text{window (h/day)}} = \frac{7.85}{10} = 0.785 \text{kW}$$

The inverter considered with this wind turbine of efficiency 90%, therefore the DC capacity becomes:

$$P\_{\text{dc,STC}} = \frac{P\_{\text{ac}} \text{(kW)}}{\eta} = \frac{0.785}{0.90} = 0.872 \text{kW}$$

For this designed house load, wind turbine capacity considered 1.1 times the load. So the adjusted wind turbine size for the equivalent DC load becomes:

$$P\_{\text{dc,STC(Adjustted)}} = 1.1 \text{xP}\_{\text{dc,STC}} = 1.1 \text{x0.872} = 0.9592 \text{kW}$$

Average wind speed of 8m/s was considered to calculate the rotor diameter for the required capacity of wind turbine. For the 0.9592kW capacity wind turbine, the rotor diameter becomes 3.95m and daily energy generated by this wind turbine was plotted in Figure 11.

Total energy generated from this hybrid system is 28.12kWh and compared with the DC load it was calculated that the Hybrid system support directly 8.45kWh of load as shown the common area in Figure 11. Therefore the minimum (18.47 - 8.45) = 10.02kWh of load needs

**Figure 11.** Hybrid system output and daily load curve shows the load on storage

to be supported by the storage system. However the hybrid system was designed to support 100% load therefore remaining generated energy (28.12 - 8.45) = 19.67kWh from hybrid system must be managed by the storage. This is the maximum load on storage.

Estimation of Energy Storage and Its Feasibility Analysis 61

**(at 24V DC system voltage)**  *Minimum (Ah) Maximum (Ah)* 

582.356Ah 1010.143Ah

543.62Ah 1067.544Ah

efficiency 12.5% with solar window 8 hours and overall wind turbine efficiency 25% with

Previous section described the required storage for the residential load in different configurations and this section describes feasibility of the storage in those configurations.

G.J. Dalton et al. in [24] compared HOMER and Hybrids as RE optimization tool and found that HOMER is better in representing hourly fluctuations in supply and demand. This chapter explains the feasibility of storage by analyzing the model output. A model was developed in HOMER version 2.68 [25] as shown in Figure 12. PV array, wind turbine, storage, inverter, grid and diesel generator were used in different size and combinations. The model identified the optimized configuration of storage, PV, wind turbine with grid or diesel generator for the residential load and investigated environmental and economical benefits due to the storage systems. The model was evaluated considering the project life time of 25 years. The performance matrices considered are NPC, COE as economical factor, Renewable Fraction (RF) and greenhouse gas (GHG) emission as environmental factor. The

1.92kW wind turbine with 5.58m rotor diameter

0.9592kW wind turbine with 3.95m rotor diameter

Hybrid system 1.27kW PV with 10.16m2 PV array and

model compared in off-grid and grid connected configurations.

**Figure 12.** Model simulation in different configurations

**Table 5.** Required storage in different configurations

**4. Feasibility analysis of storage** 

**Required system capacity Required Storage** 

2.541kW PV with 20.328m2 PV array 665.90Ah 1409.64Ah

wind window 10 hours.

**Designed system** 

Solar PV system

Wind turbine system

Considering the DC system voltage as 24V, load on battery in Ah can be calculated for one day as:

> 3 Daily minimum load in Ah @ system voltage = Load (Wh/day) 10.02x10 = = = 417.50Ah/d System Voltage 24 3 Daily maximum load in Ah @ system voltage = Load (Wh/day) 19.67x10 = = = 819.58Ah/d System Voltage 24

Considered MDOD of Lead-Acid batteries is 80%, therefore for one day discharge the battery needs to store the energy as:

> Battery storage (minimum) = Load (Ah/day) x No of days 417.50x1 <sup>=</sup> = = 521.875Ah MDOD 0.80 Battery storage (maximum) = Load (Ah/day) x No of days 819.875x1 <sup>=</sup> = = 1024.843Ah MDOD 0.80

Temperature effects are considered in the storage requirements for the hybrid systems. At 25οC for C/20 (i.e. 20 hours discharge) the discharge rate is 96%, therefore finally required battery capacity becomes:

> Required minimum Battery storage(25 C,20hour-rate)= Battery storage 521.875 <sup>=</sup> 543.62 Rated capacity 0.96 ° *Ah* Required maximum Battery storage(25 C,20hour-rate)= Battery storage 1024.843 <sup>=</sup> 1067.544 Rated capacity 0.96 ° *Ah*

Therefore in grid connected configuration to support 15.7kWh/day load the required minimum and maximum storage with solar PV, Wind turbine and hybrid system are shown in Table 5. Minimum storage indicates the required storage that needs to support the daily load. However PV, wind turbine or hybrid system generates more energy than the daily required therefore maximum storage is required to manage total generated energy by supporting load or supplying to the grid in suitable time. This design considered PV array


efficiency 12.5% with solar window 8 hours and overall wind turbine efficiency 25% with wind window 10 hours.

**Table 5.** Required storage in different configurations

## **4. Feasibility analysis of storage**

60 Energy Storage – Technologies and Applications

battery needs to store the energy as:

battery capacity becomes:

day as:

to be supported by the storage system. However the hybrid system was designed to support 100% load therefore remaining generated energy (28.12 - 8.45) = 19.67kWh from hybrid

Considering the DC system voltage as 24V, load on battery in Ah can be calculated for one

Daily minimum load in Ah @ system voltage =

Daily maximum load in Ah @ system voltage =

Load (Wh/day) 19.67x10 = = = 819.58Ah/d System Voltage 24

Considered MDOD of Lead-Acid batteries is 80%, therefore for one day discharge the

Load (Ah/day) x No of days 417.50x1 <sup>=</sup> = = 521.875Ah MDOD 0.80

Load (Ah/day) x No of days 819.875x1 <sup>=</sup> = = 1024.843Ah MDOD 0.80

Temperature effects are considered in the storage requirements for the hybrid systems. At 25οC for C/20 (i.e. 20 hours discharge) the discharge rate is 96%, therefore finally required

Required minimum Battery storage(25 C,20hour-rate)=

*Ah*

Required maximum Battery storage(25 C,20hour-rate)=

Therefore in grid connected configuration to support 15.7kWh/day load the required minimum and maximum storage with solar PV, Wind turbine and hybrid system are shown in Table 5. Minimum storage indicates the required storage that needs to support the daily load. However PV, wind turbine or hybrid system generates more energy than the daily required therefore maximum storage is required to manage total generated energy by supporting load or supplying to the grid in suitable time. This design considered PV array

*Ah*

°

°

Battery storage (minimum) =

Battery storage (maximum) =

Battery storage 521.875 <sup>=</sup> 543.62

Battery storage 1024.843 <sup>=</sup> 1067.544

Rated capacity 0.96

Rated capacity 0.96

Load (Wh/day) 10.02x10 = = = 417.50Ah/d System Voltage 24

3

3

system must be managed by the storage. This is the maximum load on storage.

Previous section described the required storage for the residential load in different configurations and this section describes feasibility of the storage in those configurations.

G.J. Dalton et al. in [24] compared HOMER and Hybrids as RE optimization tool and found that HOMER is better in representing hourly fluctuations in supply and demand. This chapter explains the feasibility of storage by analyzing the model output. A model was developed in HOMER version 2.68 [25] as shown in Figure 12. PV array, wind turbine, storage, inverter, grid and diesel generator were used in different size and combinations. The model identified the optimized configuration of storage, PV, wind turbine with grid or diesel generator for the residential load and investigated environmental and economical benefits due to the storage systems. The model was evaluated considering the project life time of 25 years. The performance matrices considered are NPC, COE as economical factor, Renewable Fraction (RF) and greenhouse gas (GHG) emission as environmental factor. The model compared in off-grid and grid connected configurations.

**Figure 12.** Model simulation in different configurations

**Net Present Cost (NPC):** The total net present cost of a system is the present value of all costs that it incurs over its lifetime minus the present value of all the revenue that it earns over its lifetime. NPC is the main economic output and ranked all systems accordingly. NPC can be represented [24-25] by Equation 10:

$$\text{NPC(\\$)}=\frac{\text{TAC}}{\text{CRF}}\tag{10}$$

Estimation of Energy Storage and Its Feasibility Analysis 63

(15)

f��� <sup>=</sup> ��������� ���������

where Eren is renewable electric production, Hren is renewable thermal production, Etot is total

**Battery dispatch:** Ideally battery charging should be taken into account for future load supply. Battery dispatch strategies explained by Barley and Winn [27], named 'load following' and 'cycle charging'. Under load following strategy, a generator produces only enough power to serve the load, which does not charge the battery. Under cycle charging strategy, whenever a generator operates it runs at its maximum rated capacity, charging battery bank with any excess electricity until the battery reaches specified state of charge. Load following strategy was considered for this analysis for the better utilization of RE.

For the simulation of the optimization model, residential load data were considered as Rockhampton resident's average load consumption. Solar radiation and wind speed data were collected from [28]. All required system components are discussed in the following

Daily average steady state load of a 3 bed room house was estimated in section 2.1 which is 15.7kWh/d. Daily load profile of the distribution network of Capricornia region was collected from [1] and according to the electricity bill information [29] the daily average electricity consumption per house is 15.7kWh/day. Model takes a set of 24 hourly values load data or monthly average or hourly load data set of 8,760 values to represent average electric load, therefore yearly residential load (AC) becomes 5730kWh/yr. The load profile is

Solar radiation data is the input to the model and hourly solar radiation data of Rockhampton was collected from [28]. Daily average extractable energy from this solar

Three hourly wind speed data was collected from [28] which was interpolated to generate hourly data and used as input data in the model. The extractable energy from the wind

For this analysis Trojan L16P Battery (6V, 360Ah) at system voltage of 24V DC is used in the

model. The efficiency of this battery is 85%, min State of Charge (SoC) 30%.

electrical production and Htot is total thermal production.

**4.1. Data collection** 

sub-sections.

*4.1.1. Electric load* 

shown in Figure 4.

*4.1.2. Solar radiation data* 

*4.1.3. Wind speed data* 

*4.1.4. Storage* 

radiation is explained in section 2.2.

speed is explained in section 2.3.

where *TAC* is the total annualized cost (which is the sum of the annualized costs of each system component). The capital recovery factor (CRF) is given by Equation 11:

$$\text{CRF} = \frac{\mathbf{i}^{\left(1+\mathbf{i}\right)^{\mathbf{N}}}}{\left(\mathbf{i}+\mathbf{i}\right)^{\mathbf{N}}-1} \tag{11}$$

where *N* denotes number of years and *i* means annual real interest rate (%). Model considered annual interest rate rather than the nominal interest rate. The overall annual interest rate considered as 6%.

**Cost of Energy (COE):** It is the average cost per kWh of useful electrical energy produced by the system. COE can be calculated by dividing the annualized cost of electricity production by the total useful electric energy production and represented [25] in Equation 12:

$$\text{COE} = \frac{\text{C}\_{\text{ann,tot}} - \text{C}\_{\text{bollower}}\text{E}\_{\text{thermal}}}{\text{E}\_{\text{prím,AC}} + \text{E}\_{\text{prím,DC}} + \text{E}\_{\text{def}} + \text{E}\_{\text{gríd,sales}}} \tag{12}$$

where Cann,tot is total annualized cost of the system (\$/yr), Cboiler is boiler marginal cost (\$/kWh), Ethermal is total thermal load served (kWh/yr), Eprim,AC is AC primary load served (kWh/yr), Eprim,DC is DC primary load served (kWh/yr), Edef is deferrable load served (kWh/yr) and Egrid,sales is total grid sales (kWh/yr).

**Emission:** Emission is widely accepted and understood environmental index. Greenhouse gases (CO2, CH4, N2O, HFCs, PFCs, SF6) are the main concern for global warming. In addition SO2 is another pollutant gas released by coal fired energy system. Emission is measured as yearly emission of the emitted gases in kg/year and emissions per capita in kg/kWh. Model used it as input when calculating the other O&M cost. It was represented in [25] as shown in Equation 13:

$$\mathbb{C}\_{\text{om,other}} = \mathbb{C}\_{\text{om,fixed}} + \mathbb{C}\_{\text{cs}} + \mathbb{C}\_{\text{emission}} \tag{13}$$

where Com,fixed is system fixed O&M cost (\$/yr), Ccs is the penalty for capacity shortage (\$/yr) and Cemission is the penalty for emission (\$/yr).

**Renewab**l**e Fraction (RF):** It is the total annual renewable power production divided by the total energy production. RF can be calculated [26] using Equation 14:

$$\mathbf{f\_{pv}} = \frac{\mathbf{E\_{PV}}}{\mathbf{E\_{TOT}}} \tag{14}$$

where EPV and ETOT are the energy generated by RE and total energy generated respectively. The overall RF (fren) can also be expressed [25] in Equation 15:

Estimation of Energy Storage and Its Feasibility Analysis 63

$$\mathbf{f\_{ren}} = \frac{\mathbf{E\_{ren}} + \mathbf{H\_{ren}}}{\mathbf{E\_{tot}} + \mathbf{H\_{tot}}} \tag{15}$$

where Eren is renewable electric production, Hren is renewable thermal production, Etot is total electrical production and Htot is total thermal production.

**Battery dispatch:** Ideally battery charging should be taken into account for future load supply. Battery dispatch strategies explained by Barley and Winn [27], named 'load following' and 'cycle charging'. Under load following strategy, a generator produces only enough power to serve the load, which does not charge the battery. Under cycle charging strategy, whenever a generator operates it runs at its maximum rated capacity, charging battery bank with any excess electricity until the battery reaches specified state of charge. Load following strategy was considered for this analysis for the better utilization of RE.

## **4.1. Data collection**

62 Energy Storage – Technologies and Applications

can be represented [24-25] by Equation 10:

interest rate considered as 6%.

[25] as shown in Equation 13:

(kWh/yr) and Egrid,sales is total grid sales (kWh/yr).

and Cemission is the penalty for emission (\$/yr).

**Net Present Cost (NPC):** The total net present cost of a system is the present value of all costs that it incurs over its lifetime minus the present value of all the revenue that it earns over its lifetime. NPC is the main economic output and ranked all systems accordingly. NPC

NPC(\$) <sup>=</sup> ���

where *TAC* is the total annualized cost (which is the sum of the annualized costs of each

CRF = �(���)�

where *N* denotes number of years and *i* means annual real interest rate (%). Model considered annual interest rate rather than the nominal interest rate. The overall annual

**Cost of Energy (COE):** It is the average cost per kWh of useful electrical energy produced by the system. COE can be calculated by dividing the annualized cost of electricity production

where Cann,tot is total annualized cost of the system (\$/yr), Cboiler is boiler marginal cost (\$/kWh), Ethermal is total thermal load served (kWh/yr), Eprim,AC is AC primary load served (kWh/yr), Eprim,DC is DC primary load served (kWh/yr), Edef is deferrable load served

**Emission:** Emission is widely accepted and understood environmental index. Greenhouse gases (CO2, CH4, N2O, HFCs, PFCs, SF6) are the main concern for global warming. In addition SO2 is another pollutant gas released by coal fired energy system. Emission is measured as yearly emission of the emitted gases in kg/year and emissions per capita in kg/kWh. Model used it as input when calculating the other O&M cost. It was represented in

where Com,fixed is system fixed O&M cost (\$/yr), Ccs is the penalty for capacity shortage (\$/yr)

**Renewab**l**e Fraction (RF):** It is the total annual renewable power production divided by the

f�� <sup>=</sup> ��� ����

where EPV and ETOT are the energy generated by RE and total energy generated respectively.

total energy production. RF can be calculated [26] using Equation 14:

The overall RF (fren) can also be expressed [25] in Equation 15:

C�������� = C�������� + C�� + C�������� (13)

(14)

����������������������������������

by the total useful electric energy production and represented [25] in Equation 12:

COE = ������������������������

system component). The capital recovery factor (CRF) is given by Equation 11:

��� (10)

(���)��� (11)

(12)

For the simulation of the optimization model, residential load data were considered as Rockhampton resident's average load consumption. Solar radiation and wind speed data were collected from [28]. All required system components are discussed in the following sub-sections.

## *4.1.1. Electric load*

Daily average steady state load of a 3 bed room house was estimated in section 2.1 which is 15.7kWh/d. Daily load profile of the distribution network of Capricornia region was collected from [1] and according to the electricity bill information [29] the daily average electricity consumption per house is 15.7kWh/day. Model takes a set of 24 hourly values load data or monthly average or hourly load data set of 8,760 values to represent average electric load, therefore yearly residential load (AC) becomes 5730kWh/yr. The load profile is shown in Figure 4.

### *4.1.2. Solar radiation data*

Solar radiation data is the input to the model and hourly solar radiation data of Rockhampton was collected from [28]. Daily average extractable energy from this solar radiation is explained in section 2.2.

## *4.1.3. Wind speed data*

Three hourly wind speed data was collected from [28] which was interpolated to generate hourly data and used as input data in the model. The extractable energy from the wind speed is explained in section 2.3.

## *4.1.4. Storage*

For this analysis Trojan L16P Battery (6V, 360Ah) at system voltage of 24V DC is used in the model. The efficiency of this battery is 85%, min State of Charge (SoC) 30%.

## **4.2. System components cost**

Table 6 lists the required system components with related costs in Australian currency. PV array, Wind turbine, Battery charger, Inverter, deep cycle battery, diesel generator and grid electricity costs are included for the analysis. PV array including inverter price is available, and found that 1.52kW PV array with inverter costs \$3599 [30], also it is found that 1.56kW PV with inverter costs is \$4991[31]. However model considered battery charger is included with PV array therefore the PV array cost is listed accordingly in Table 6 and inverter costs considered separately.

Estimation of Energy Storage and Its Feasibility Analysis 65

Description Value/Information

SMA Sunny Boy Grid Tie Inverter (7000Watt SB7000US) price is \$2823 [32], however Sunny Boy 1700W inverter price is \$699 [33]. 1kW BWC XL.1 wind turbine with 24V DC charge controller price is \$3560 [34] and 10kW Bergey BWC Excel with battery charging or grid tied option wind turbine cost is \$29,250 [35]. Grid electricity cost in Rockhampton is found from Ergon Energy's electricity bill [36] and for Tariff-11, it is \$0.285/kWh (including GST & service). However Government's decision to impose carbon tax at the rate of \$23/ton of GHG emission which will increase this electricity bill as well as the cost of conventional energy sources, therefore off-peak electricity cost is considered as \$0.30/kWh for analysis. Trojan T-105 6V, 225AH (20HR) Flooded Lead Acid Battery price is \$124.79 [37]. Fuel cost

for generator is considered at the current price available in Rockhampton, Australia.

The significance of storage was analyzed from the optimized model to evaluate environmental and economical advantages of storage in off-grid and grid-connected configurations in fourteen different cases. All these cases were analyzed considering same

Capital cost \$2200.00/kW Replacement cost \$2000.00/kW Operation & maintenance cost \$0.05/hr Life time 15000hrs Fuel cost \$1.53/ltr

**Table 6.** Technical Data and Study assumptions

load 15.7kWh/d or 5730kWh/yr.

Case-1: Diesel Generator only Case-2: PV with Diesel Generator

Case-1: Grid only

**Category-1: Off-grid Configuration** 

Case-3: PV with Storage and Diesel Generator Case-4: Wind turbine with Diesel Generator

**Category-2: Grid-connected Configuration** 

Case-2: PV with Grid and Diesel generator

Case-3: PV with Storage, Grid and Diesel generator Case-4: Wind turbine with Grid and Diesel generator

Case-5: Wind turbine with Storage, Grid and Diesel generator Case-6: Hybrid (PV & Wind turbine) with Grid and Diesel generator

Case-5: Wind turbine with Storage and Diesel Generator Case-6: Hybrid (PV & Wind turbine) with Diesel Generator

Case-7: Hybrid (PV & Wind turbine) with Storage and Diesel Generator

Case-7: Hybrid (PV & Wind turbine) with Storage, Grid and Diesel generator

Generator



**Table 6.** Technical Data and Study assumptions

64 Energy Storage – Technologies and Applications

Wind Turbine (BWC XL.1 1 kW DC)

Table 6 lists the required system components with related costs in Australian currency. PV array, Wind turbine, Battery charger, Inverter, deep cycle battery, diesel generator and grid electricity costs are included for the analysis. PV array including inverter price is available, and found that 1.52kW PV array with inverter costs \$3599 [30], also it is found that 1.56kW PV with inverter costs is \$4991[31]. However model considered battery charger is included with PV array therefore the PV array cost is listed accordingly in Table 6 and inverter costs

Description Value/Information

Capital cost \$3100.00/kW Replacement cost \$3000.00/kW Life Time 25 years Operation & maintenance cost \$50.00/year

Capacity 1kW DC Hub Height 40m Capital cost \$4000.00 Replacement cost \$3000.00 Life time 25 years Operation & maintenance cost \$120/yr

Electricity price (Off peak time) \$0.30/kWh Electricity price (Peak time) \$.42/kWh Electricity price (Super Peak time) \$0.75/kWh

CO2 632.0 g/kWh CO 0.7 g/kWh Unburned hydrocarbons 0.08 g/kWh Particulate matter 0.052 g/kWh SO2 2.74 g/kWh NOx 1.34 g/kWh

Capital cost \$400.00/kW Replacement cost \$325.00/kW Life time 15 years Operation & maintenance cost \$25.00/year

Capital cost \$170.00/6V 360Ah Replacement cost \$130.00/6V 360Ah

System Voltage 24 volts

**4.2. System components cost** 

considered separately.

PV array

Grid electricity

Emission factor

Inverter

Storage (Battery)

SMA Sunny Boy Grid Tie Inverter (7000Watt SB7000US) price is \$2823 [32], however Sunny Boy 1700W inverter price is \$699 [33]. 1kW BWC XL.1 wind turbine with 24V DC charge controller price is \$3560 [34] and 10kW Bergey BWC Excel with battery charging or grid tied option wind turbine cost is \$29,250 [35]. Grid electricity cost in Rockhampton is found from Ergon Energy's electricity bill [36] and for Tariff-11, it is \$0.285/kWh (including GST & service). However Government's decision to impose carbon tax at the rate of \$23/ton of GHG emission which will increase this electricity bill as well as the cost of conventional energy sources, therefore off-peak electricity cost is considered as \$0.30/kWh for analysis. Trojan T-105 6V, 225AH (20HR) Flooded Lead Acid Battery price is \$124.79 [37]. Fuel cost for generator is considered at the current price available in Rockhampton, Australia.

The significance of storage was analyzed from the optimized model to evaluate environmental and economical advantages of storage in off-grid and grid-connected configurations in fourteen different cases. All these cases were analyzed considering same load 15.7kWh/d or 5730kWh/yr.

#### **Category-1: Off-grid Configuration**

Case-1: Diesel Generator only Case-2: PV with Diesel Generator Case-3: PV with Storage and Diesel Generator Case-4: Wind turbine with Diesel Generator Case-5: Wind turbine with Storage and Diesel Generator Case-6: Hybrid (PV & Wind turbine) with Diesel Generator Case-7: Hybrid (PV & Wind turbine) with Storage and Diesel Generator

#### **Category-2: Grid-connected Configuration**

Case-1: Grid only Case-2: PV with Grid and Diesel generator Case-3: PV with Storage, Grid and Diesel generator Case-4: Wind turbine with Grid and Diesel generator Case-5: Wind turbine with Storage, Grid and Diesel generator Case-6: Hybrid (PV & Wind turbine) with Grid and Diesel generator Case-7: Hybrid (PV & Wind turbine) with Storage, Grid and Diesel generator

## **5. Results and discussion**

Simulation was conducted to get optimized configuration of RE resources. Simulation results and findings are discussed below.

Estimation of Energy Storage and Its Feasibility Analysis 67

**Case 4. Wind turbine with Diesel Generator configuration** 

minimized and that could reduce the use of diesel generator.

total generated energy was wasted that could be sold to the grid.

**Case 6. Hybrid system with Diesel generator configuration** 

demand for 24 hours period.

of diesel generator.

**Case 5. Wind turbine with Storage and Diesel Generator configuration** 

load demand and battery stored the excess electricity to support at other time.

meet the load demand for 24 hours period.

This off-grid configuration used 10kW BWC XL.1 wind generator with 5kW inverter and 10kW diesel generator as required resources to support 5730kWh/yr of load. Result showed that, wind turbine generates much more electricity than the total load demand but could not

Total 41,023kWh/yr of electricity was generated from wind turbine and diesel generator, where 38,781kWh/yr from wind turbine i.e. 94.5% of total production from RE but most of it was wasted as wind turbine supports 3488kWh/yr of load, which is 60.87% of load demand. Diesel generator contributes 2242kWh/yr or 39.13% of load demand, although compared to the total production; diesel generator contribution was only 5.5%. Total 35,070kWh/yr or 85.5% of total electricity production was wasted but compared to the total wind turbine output 90.43% was wasted. By adding storage this huge loss of electricity could be

This off-grid configuration model used 3kW BWC XL.1 wind generator, 40 numbers of Trojan L16P Battery (@ 6V, 360Ah) at 24V system voltage and 5kW Inverter. This optimized configuration shaded out diesel generator, therefore 100% load was supported by wind turbine and storage. Result showed that, wind turbine generates electricity more than the

Wind turbine generates 11,634kWh/yr of electricity. Inverter converts 6096kWh/yr of DC electricity to AC. Battery stored 2364kWh/yr of energy and supplied 2037kWh/yr to the load. Battery stored 20.32% of wind turbine generated energy and supported 35.55% of load. However still 5211kWh/yr of excess energy generated by the wind turbine, i.e 44.79% of

This off-grid hybrid configuration model used 3kW PV, 5kW BWC XL.1 wind generator, 5kW Inverter and 4kW diesel generator. Result showed that, although PV and wind turbine generates much more electricity than the total load demand but could not meet the load

Total electricity generated from RE (PV and Wind turbine) and diesel generator was 24,434kWh/yr where 2,227kWh/yr from PV, 19,390kWh/yr from wind turbine and 2,817kWh/yr from diesel generator. PV contributed 9.1%, wind turbine 79.35% and diesel generator 11.53% of total production, therefore overall RE contribution was 88.5% of total production. Diesel generator contributed 49.16% of load demand. Inverter converts 3,099kWh/yr of DC electricity to 2,913kWh/yr of AC electricity from RE generation which was 50.84% of load demand. A significant amount of electricity (18,518kWh/yr) from RE was wasted which is 75.78% of total electricity production and 85.66% compared to the total RE production. Storage could be used to reduce this huge energy loss and to minimize the use

## **5.1. Category - 1(Off grid configuration)**

## **Case 1. Diesel Generator only**

In this configuration 10kW Diesel generator was used to support total load of 5730kWh/yr which consumed enough fuel (8440L/yr) and emitted significant amount of GHG & pollutant gas to the air. Generator required frequent maintenance and fuel cost was also high therefore NPC was high and COE was \$5.342/kWh. This configuration was the costliest and environmentally most vulnerable.

## **Case 2. PV with Diesel Generator configuration**

In this off-grid configuration, model used 12kW PV with 5kW Inverter and 10kW Diesel generator as required resources. Results showed that, although PV generates electricity more than the total load demand but could not meet the load demand during night. Total 12,781kWh/yr electricity was generated from PV and diesel generator. PV alone generates 8908kWh/yr i.e. RF became 69.7% but most of the energy from PV array was wasted. Diesel generator directly supplied 3873kWh/yr to the load which was 67.6% of load demand, although compared to the total production; generator contribution was only 30.3%. The remaining load demand, (5730 -3873) = 1857kWh/yr was supported by PV array through inverter. Therefore a significant amount of electricity from PV array was wasted. Wasted electricity was (8908 - 1857/0.94) = 6932.46kWh/yr which is 54.2% of total electricity production but compared to the total PV electricity production, the wasted electricity was 77.82%. To reduce this great amount of loss, this system should have some way to store the energy and could reduce the use of diesel generator.

### **Case 3. PV with Storage and Diesel Generator configuration**

In this off-grid configuration model 11kW PV, 48 number of Trojan L16P Battery (@ 6V, 360Ah) at 24V system voltage with 5kW Inverter was used. The optimized configuration shaded out diesel generator, therefore 100% load supported by PV and storage. Results showed that, PV generates electricity more than the load demand and battery stored the excess electricity to maintain the load demand.

PV generates 8166kWh/yr of electricity from which a good amount of energy was stored in the battery and used at other time. Total AC load supported directly by PV array during day time and by battery during morning & night. Inverter converts 6096kWh/yr of DC electricity to AC. Battery stored 4281kWh/yr of energy and supplied 3692kWh/yr to support the load. Battery stored 52.42% of PV generated energy and supported 64.43% of load while PV directly supports 35.56% of load. However still 1480kWh/yr of excess energy generated by the PV array and was wasted that could be sold to the grid. This model configuration supports 100% load by PV array and storage which makes it environment friendly off-grid configuration.

## **Case 4. Wind turbine with Diesel Generator configuration**

66 Energy Storage – Technologies and Applications

results and findings are discussed below.

and environmentally most vulnerable.

**Case 2. PV with Diesel Generator configuration** 

energy and could reduce the use of diesel generator.

excess electricity to maintain the load demand.

**Case 3. PV with Storage and Diesel Generator configuration** 

**5.1. Category - 1(Off grid configuration)** 

Simulation was conducted to get optimized configuration of RE resources. Simulation

In this configuration 10kW Diesel generator was used to support total load of 5730kWh/yr which consumed enough fuel (8440L/yr) and emitted significant amount of GHG & pollutant gas to the air. Generator required frequent maintenance and fuel cost was also high therefore NPC was high and COE was \$5.342/kWh. This configuration was the costliest

In this off-grid configuration, model used 12kW PV with 5kW Inverter and 10kW Diesel generator as required resources. Results showed that, although PV generates electricity more than the total load demand but could not meet the load demand during night. Total 12,781kWh/yr electricity was generated from PV and diesel generator. PV alone generates 8908kWh/yr i.e. RF became 69.7% but most of the energy from PV array was wasted. Diesel generator directly supplied 3873kWh/yr to the load which was 67.6% of load demand, although compared to the total production; generator contribution was only 30.3%. The remaining load demand, (5730 -3873) = 1857kWh/yr was supported by PV array through inverter. Therefore a significant amount of electricity from PV array was wasted. Wasted electricity was (8908 - 1857/0.94) = 6932.46kWh/yr which is 54.2% of total electricity production but compared to the total PV electricity production, the wasted electricity was 77.82%. To reduce this great amount of loss, this system should have some way to store the

In this off-grid configuration model 11kW PV, 48 number of Trojan L16P Battery (@ 6V, 360Ah) at 24V system voltage with 5kW Inverter was used. The optimized configuration shaded out diesel generator, therefore 100% load supported by PV and storage. Results showed that, PV generates electricity more than the load demand and battery stored the

PV generates 8166kWh/yr of electricity from which a good amount of energy was stored in the battery and used at other time. Total AC load supported directly by PV array during day time and by battery during morning & night. Inverter converts 6096kWh/yr of DC electricity to AC. Battery stored 4281kWh/yr of energy and supplied 3692kWh/yr to support the load. Battery stored 52.42% of PV generated energy and supported 64.43% of load while PV directly supports 35.56% of load. However still 1480kWh/yr of excess energy generated by the PV array and was wasted that could be sold to the grid. This model configuration supports 100% load by PV array and storage which makes it environment friendly off-grid configuration.

**5. Results and discussion** 

**Case 1. Diesel Generator only** 

This off-grid configuration used 10kW BWC XL.1 wind generator with 5kW inverter and 10kW diesel generator as required resources to support 5730kWh/yr of load. Result showed that, wind turbine generates much more electricity than the total load demand but could not meet the load demand for 24 hours period.

Total 41,023kWh/yr of electricity was generated from wind turbine and diesel generator, where 38,781kWh/yr from wind turbine i.e. 94.5% of total production from RE but most of it was wasted as wind turbine supports 3488kWh/yr of load, which is 60.87% of load demand. Diesel generator contributes 2242kWh/yr or 39.13% of load demand, although compared to the total production; diesel generator contribution was only 5.5%. Total 35,070kWh/yr or 85.5% of total electricity production was wasted but compared to the total wind turbine output 90.43% was wasted. By adding storage this huge loss of electricity could be minimized and that could reduce the use of diesel generator.

## **Case 5. Wind turbine with Storage and Diesel Generator configuration**

This off-grid configuration model used 3kW BWC XL.1 wind generator, 40 numbers of Trojan L16P Battery (@ 6V, 360Ah) at 24V system voltage and 5kW Inverter. This optimized configuration shaded out diesel generator, therefore 100% load was supported by wind turbine and storage. Result showed that, wind turbine generates electricity more than the load demand and battery stored the excess electricity to support at other time.

Wind turbine generates 11,634kWh/yr of electricity. Inverter converts 6096kWh/yr of DC electricity to AC. Battery stored 2364kWh/yr of energy and supplied 2037kWh/yr to the load. Battery stored 20.32% of wind turbine generated energy and supported 35.55% of load. However still 5211kWh/yr of excess energy generated by the wind turbine, i.e 44.79% of total generated energy was wasted that could be sold to the grid.

### **Case 6. Hybrid system with Diesel generator configuration**

This off-grid hybrid configuration model used 3kW PV, 5kW BWC XL.1 wind generator, 5kW Inverter and 4kW diesel generator. Result showed that, although PV and wind turbine generates much more electricity than the total load demand but could not meet the load demand for 24 hours period.

Total electricity generated from RE (PV and Wind turbine) and diesel generator was 24,434kWh/yr where 2,227kWh/yr from PV, 19,390kWh/yr from wind turbine and 2,817kWh/yr from diesel generator. PV contributed 9.1%, wind turbine 79.35% and diesel generator 11.53% of total production, therefore overall RE contribution was 88.5% of total production. Diesel generator contributed 49.16% of load demand. Inverter converts 3,099kWh/yr of DC electricity to 2,913kWh/yr of AC electricity from RE generation which was 50.84% of load demand. A significant amount of electricity (18,518kWh/yr) from RE was wasted which is 75.78% of total electricity production and 85.66% compared to the total RE production. Storage could be used to reduce this huge energy loss and to minimize the use of diesel generator.

## **Case 7. Hybrid system with Storage and Diesel Generator configuration**

This off-grid hybrid configuration model used 1kW PV, 3kW BWC XL.1 wind generator, 5kW Inverter and 32 Trojan L16P batteries at 24V system voltage to support the same load. Hybrid system (PV and wind turbine) output with storage supports 100% load and shaded out the use of diesel generator. Result showed that, storage managed the electricity from hybrid system and met the load demand 24 hours a day, but a significant amount of energy was wasted that could be sold to the grid.

Estimation of Energy Storage and Its Feasibility Analysis 69

Load support

(103.68kWh) 100% 18.12%

(86.4kWh) 100% 44.79%

(69.12kWh) 100% 48.44%

RE energy loss

**PV Wind Inverter Storage RE use** 

system voltage)

(at 24V DC

(PV+Storage+Gen) 11kW - 5kW 48 nos.

(Hybrid+Storage+Gen) 1kW 3kW 5kW 32 nos.

**Case 3. PV with Storage in Grid connected configuration** 

**Case 4. Wind turbine in Grid connected configuration** 

to the grid. Total 894kWh/yr of electricity was unused.

**Table 7.** Category-1 or off-grid configuration results

(PV +Gen) 12kW - 5kW - 32.40% 77.82%

(Wind turbine+Gen) - 10kW 5kW - 60.87% 90.43%

(Hybrid +Gen) 3kW 5kW 5kW - 50.84% 85.66%

This configuration model is very interesting compared to the earlier case that, by adding sufficient amount of storage, system improved PV contribution for same load demand. To meet load demand this model used 5kW PV, 12 numbers of Trojan L16P battery, 5kW inverter and grid supply. Total 6208kWh/yr of electricity produced where grid supplied 2496kWh/yr or 40% of total production or 43.56% of total load demand. PV array produced 3712kWh/yr or 60% of total production. Loss of energy was very insignificant. Battery stored 1918kWh/yr and supplied 1648kWh/yr to the load or 28.76% of total load. However PV array directly supported (5730-2496-1648) = 1586kWh/yr of load which was 27.68% of total

In this configuration model, wind turbine generates enough electricity but was unable to meet the timely load demand therefore consumed sufficient amount of grid electricity. To meet the load demand this optimized model used 3kW BWC XL.1 wind turbine, 5kW inverter and grid supply. Total 14,389kWh/yr of electricity was produced where wind generator produced 11,634kWh/yr or 80.9% of total production. Grid supplied 2755kWh/yr or 19.1% of total production or 48.08% of load demand. Wind turbine supported (5730 - 2755) = 2975kWh/yr or 51.92% of load demand and 7121kWh/yr of electricity was sold back


Case-2

Case-3:

Case-4

Case-5 (Wind

Case-6

Case-7

load demand.

turbine+Storage+Gen)

Total 12,376kWh/yr of electricity was generated from hybrid system, where 742kWh/yr or 6% from PV and 11,634kWh/yr or 94% from wind turbine. Battery stored 2073kWh/yr and supported 1,788kWh/yr or 31.20% of load demand. This configuration supplied 100% load demand from RE, however 5,995kWh/yr was wasted which could be sold to the grid.

## **Summary of Category-1 or Off-grid configurations**

The results of standalone configurations can be summarized that storage minimized the use of resources which reduced the project cost, improved load support that reduced GHG emission and reduced the loss of generated RE and showed the scope to sell excess energy to the grid. Table 7 summarizes these findings. Load support describes, percentage of load supported by RE and storage. Energy loss describes percentage of energy loss compared to total RE production.

## **5.2. Category-2 (Grid connected configuration)**

## **Case 1. Grid only configuration**

This is the present configuration of most residential electricity connection. Grid supplies total load demand of 5730kWh/yr. Grid electricity tariff varies with time, season and application [36, 38]. This configuration model considered 3 different price of grid electricity, depending on demand time. These are off-peak (0.30\$/kWh), peak (0.42\$/kWh) and super peak rate (0.75\$/kWh). 6:00PM to 7:00PM considered super peak, 8:00PM to 10:00PM and 8:00AM to 9:00AM considered peak time and rest are off peak time. In this case yearly average COE becomes \$0.422/kWh. As grid electricity mainly comes from conventional sources therefore a good amount of GHG and pollutant gas emits to the air.

### **Case 2. PV in Grid connected configuration**

In this optimized model configuration diesel Generator was shaded out, however PV array still contributed a small portion of load demand. To meet load demand this model used 3kW PV, 5kW Inverter and grid supply. Total 7,182kWh/yr electricity was produced, where grid supplied 4,955kWh/yr or 69% of total production or 86.47% of total load demand. PV array produced 2,227kWh/yr or 31% of total production or 13.53% of the load demand. Total 549kWh/yr of energy was sold back to the grid and 818kWh/yr of PV generated electricity was wasted due to mismatch in timely demand which could be stored and supplied to the load.


**Table 7.** Category-1 or off-grid configuration results

68 Energy Storage – Technologies and Applications

was wasted that could be sold to the grid.

total RE production.

load.

**Case 1. Grid only configuration** 

**Summary of Category-1 or Off-grid configurations** 

**5.2. Category-2 (Grid connected configuration)** 

**Case 2. PV in Grid connected configuration** 

**Case 7. Hybrid system with Storage and Diesel Generator configuration** 

This off-grid hybrid configuration model used 1kW PV, 3kW BWC XL.1 wind generator, 5kW Inverter and 32 Trojan L16P batteries at 24V system voltage to support the same load. Hybrid system (PV and wind turbine) output with storage supports 100% load and shaded out the use of diesel generator. Result showed that, storage managed the electricity from hybrid system and met the load demand 24 hours a day, but a significant amount of energy

Total 12,376kWh/yr of electricity was generated from hybrid system, where 742kWh/yr or 6% from PV and 11,634kWh/yr or 94% from wind turbine. Battery stored 2073kWh/yr and supported 1,788kWh/yr or 31.20% of load demand. This configuration supplied 100% load

The results of standalone configurations can be summarized that storage minimized the use of resources which reduced the project cost, improved load support that reduced GHG emission and reduced the loss of generated RE and showed the scope to sell excess energy to the grid. Table 7 summarizes these findings. Load support describes, percentage of load supported by RE and storage. Energy loss describes percentage of energy loss compared to

This is the present configuration of most residential electricity connection. Grid supplies total load demand of 5730kWh/yr. Grid electricity tariff varies with time, season and application [36, 38]. This configuration model considered 3 different price of grid electricity, depending on demand time. These are off-peak (0.30\$/kWh), peak (0.42\$/kWh) and super peak rate (0.75\$/kWh). 6:00PM to 7:00PM considered super peak, 8:00PM to 10:00PM and 8:00AM to 9:00AM considered peak time and rest are off peak time. In this case yearly average COE becomes \$0.422/kWh. As grid electricity mainly comes from conventional

In this optimized model configuration diesel Generator was shaded out, however PV array still contributed a small portion of load demand. To meet load demand this model used 3kW PV, 5kW Inverter and grid supply. Total 7,182kWh/yr electricity was produced, where grid supplied 4,955kWh/yr or 69% of total production or 86.47% of total load demand. PV array produced 2,227kWh/yr or 31% of total production or 13.53% of the load demand. Total 549kWh/yr of energy was sold back to the grid and 818kWh/yr of PV generated electricity was wasted due to mismatch in timely demand which could be stored and supplied to the

sources therefore a good amount of GHG and pollutant gas emits to the air.

demand from RE, however 5,995kWh/yr was wasted which could be sold to the grid.

## **Case 3. PV with Storage in Grid connected configuration**

This configuration model is very interesting compared to the earlier case that, by adding sufficient amount of storage, system improved PV contribution for same load demand. To meet load demand this model used 5kW PV, 12 numbers of Trojan L16P battery, 5kW inverter and grid supply. Total 6208kWh/yr of electricity produced where grid supplied 2496kWh/yr or 40% of total production or 43.56% of total load demand. PV array produced 3712kWh/yr or 60% of total production. Loss of energy was very insignificant. Battery stored 1918kWh/yr and supplied 1648kWh/yr to the load or 28.76% of total load. However PV array directly supported (5730-2496-1648) = 1586kWh/yr of load which was 27.68% of total load demand.

### **Case 4. Wind turbine in Grid connected configuration**

In this configuration model, wind turbine generates enough electricity but was unable to meet the timely load demand therefore consumed sufficient amount of grid electricity. To meet the load demand this optimized model used 3kW BWC XL.1 wind turbine, 5kW inverter and grid supply. Total 14,389kWh/yr of electricity was produced where wind generator produced 11,634kWh/yr or 80.9% of total production. Grid supplied 2755kWh/yr or 19.1% of total production or 48.08% of load demand. Wind turbine supported (5730 - 2755) = 2975kWh/yr or 51.92% of load demand and 7121kWh/yr of electricity was sold back to the grid. Total 894kWh/yr of electricity was unused.

## **Case 5. Wind turbine with Storage in Grid connected configuration**

This configuration model used 3kW BWC XL.1 wind generator, 16 numbers of Trojan L16P battery, 5kW inverter and grid supply. Total 11,784kWh/yr of electricity was produced where grid supplied only 150kWh/yr or 1.3% of total production or only 2.6% of total load demand. Wind turbine produced 11,634kWh/yr which was 98.7% of total production. Battery stored 2202kWh/yr and supplied 1895kWh/yr or 33.07% of total load demand. However wind turbine directly supported (5730-1895-150) = 3685kWh/yr or 64.31% of total load demand. Total 5068kWh/yr or 53.65% of wind production was sold back to the grid. Significant amount of electricity was sold back to the grid therefore overall GHG emission was reduced.

Estimation of Energy Storage and Its Feasibility Analysis 71

Load support

(25.92kWh) 56.44% 0.027% 0.0%

(34.56kWh) 97.38% 43.56% 0.0%

(25.92kWh) 97.03% 44.66% 2.53%

**RE use** 

Grid sales

RE energy loss

**Storage**  (at 24V DC system voltage)


**PV Wind Inverter** 

(PV+Storage+Grid) 5kW - 5kW 12 nos.

**Table 8.** Category-2 or grid connected configuration results

(PV +Grid) 3kW - 5kW - 38.86% 24.65% 36.73%


1kW 3kW 3kW 12 nos.

(Hybrid +Grid) 1kW 1kW 1kW - 23.19% 26.75% 40.93%

The optimization was done in two configuration categories and seven cases in each category. Four different factors were compared in each case. These factors were GHG & Pollutant gas emission, RF, COE and NPC. The comparative findings of these factors are

Figure 13 shows GHG and pollutant gas emissions in different case configurations. It was found that by adding storage in stand-alone system, emission of GHG and other pollutant gas was eliminated. In Grid connected configuration, it was also evident that storage minimized emission by improving RE utilization. By selling excess energy back to the grid storage with wind and hybrid system further helped in reducing GHG emission from grid

RF is the measuring index of how much electricity produced from RE, out of total production. In stand-alone system it was found that storage eliminates the use of diesel generator therefore RF became 100%. In Grid connected configuration Storage again

improves the RE utilization and RF became as high as 98.7% as shown in Figure 14.

Case-2

Case-3

Case-4

Grid)

Case-5

Case-6

Case-7

+Grid)

(Wind turbine +

(Wind turbine + Storage +Grid)

(Hybrid+Storage

**5.3. Findings** 

explained below.

**GHG & Pollutant gas emission** 

**Renewable Fraction (RF)** 

which is shown in negative values in Figure 13.

## **Case 6. Hybrid system without Storage in Grid connected configuration**

In this configuration both PV and wind turbine was used. This hybrid configuration model used 1kW PV, 1kW BWC XL.1 wind generator, 1kW inverter and grid supply for the same load of 5730kWh/yr. Results showed that, the hybrid system was optimized such that minimum RE components were required but could not met the load demand for 24 hours period. Total 9,021kWh/yr of electricity was produced where grid supplied 4,401kWh/yr or 76.80% of load demand or 48.78% of total production. PV and wind hybrid system produced 4,620kWh/yr or 51.21% of total production. Hybrid system supplied 1329kWh/yr of electricity or 23.19% of load demand. However hybrid system generated enough electricity and sold 1,236kWh/yr to the grid, still 1,891kWh/yr of electricity wasted which was 40.93% of total RE production. This wasted electricity could be utilized and grid use could be minimized by adding storage.

### **Case 7. Hybrid system with Storage in Grid connected configuration**

This hybrid configuration model used 1kW PV, 3kW BWC XL.1 wind generator, 3kW inverter, 12 Trojan L16P battery and grid supply. This configuration improved RE contribution in supporting load. Total 12,546kWh/yr of electricity was produced where grid supplied only 170kWh/yr which is 1.4% of total production or 2.96% of load demand. PV generates 742kWh/yr and wind turbine 11,634kWh/yr i.e. RE production was 98.6% of total electricity generation. This hybrid system sold back 5,527kWh/yr of electricity to the grid. Storage helped in improving RE utilization & minimized loss. Battery stored 1911kWh/yr and supported 1642kWh/yr of load or 28.65% of load demand.

### **Summary of Category-2 or Grid-connected configurations**

The results of grid connected configurations can be summarized that storage improved load support which reduced GHG emission. Storage optimized the RE sources by minimizing grid use and reduced loss of energy. Table 8 summarizes these findings. Load support describes, percentage of load supported by RE and storage. Energy loss describes percentage of energy loss compared to total RE production. Grid sales describes, energy sold to grid compared to total RE production.


**Table 8.** Category-2 or grid connected configuration results

## **5.3. Findings**

70 Energy Storage – Technologies and Applications

was reduced.

minimized by adding storage.

**Case 5. Wind turbine with Storage in Grid connected configuration** 

**Case 6. Hybrid system without Storage in Grid connected configuration** 

**Case 7. Hybrid system with Storage in Grid connected configuration** 

and supported 1642kWh/yr of load or 28.65% of load demand.

**Summary of Category-2 or Grid-connected configurations** 

to grid compared to total RE production.

This configuration model used 3kW BWC XL.1 wind generator, 16 numbers of Trojan L16P battery, 5kW inverter and grid supply. Total 11,784kWh/yr of electricity was produced where grid supplied only 150kWh/yr or 1.3% of total production or only 2.6% of total load demand. Wind turbine produced 11,634kWh/yr which was 98.7% of total production. Battery stored 2202kWh/yr and supplied 1895kWh/yr or 33.07% of total load demand. However wind turbine directly supported (5730-1895-150) = 3685kWh/yr or 64.31% of total load demand. Total 5068kWh/yr or 53.65% of wind production was sold back to the grid. Significant amount of electricity was sold back to the grid therefore overall GHG emission

In this configuration both PV and wind turbine was used. This hybrid configuration model used 1kW PV, 1kW BWC XL.1 wind generator, 1kW inverter and grid supply for the same load of 5730kWh/yr. Results showed that, the hybrid system was optimized such that minimum RE components were required but could not met the load demand for 24 hours period. Total 9,021kWh/yr of electricity was produced where grid supplied 4,401kWh/yr or 76.80% of load demand or 48.78% of total production. PV and wind hybrid system produced 4,620kWh/yr or 51.21% of total production. Hybrid system supplied 1329kWh/yr of electricity or 23.19% of load demand. However hybrid system generated enough electricity and sold 1,236kWh/yr to the grid, still 1,891kWh/yr of electricity wasted which was 40.93% of total RE production. This wasted electricity could be utilized and grid use could be

This hybrid configuration model used 1kW PV, 3kW BWC XL.1 wind generator, 3kW inverter, 12 Trojan L16P battery and grid supply. This configuration improved RE contribution in supporting load. Total 12,546kWh/yr of electricity was produced where grid supplied only 170kWh/yr which is 1.4% of total production or 2.96% of load demand. PV generates 742kWh/yr and wind turbine 11,634kWh/yr i.e. RE production was 98.6% of total electricity generation. This hybrid system sold back 5,527kWh/yr of electricity to the grid. Storage helped in improving RE utilization & minimized loss. Battery stored 1911kWh/yr

The results of grid connected configurations can be summarized that storage improved load support which reduced GHG emission. Storage optimized the RE sources by minimizing grid use and reduced loss of energy. Table 8 summarizes these findings. Load support describes, percentage of load supported by RE and storage. Energy loss describes percentage of energy loss compared to total RE production. Grid sales describes, energy sold The optimization was done in two configuration categories and seven cases in each category. Four different factors were compared in each case. These factors were GHG & Pollutant gas emission, RF, COE and NPC. The comparative findings of these factors are explained below.

## **GHG & Pollutant gas emission**

Figure 13 shows GHG and pollutant gas emissions in different case configurations. It was found that by adding storage in stand-alone system, emission of GHG and other pollutant gas was eliminated. In Grid connected configuration, it was also evident that storage minimized emission by improving RE utilization. By selling excess energy back to the grid storage with wind and hybrid system further helped in reducing GHG emission from grid which is shown in negative values in Figure 13.

## **Renewable Fraction (RF)**

RF is the measuring index of how much electricity produced from RE, out of total production. In stand-alone system it was found that storage eliminates the use of diesel generator therefore RF became 100%. In Grid connected configuration Storage again improves the RE utilization and RF became as high as 98.7% as shown in Figure 14.

Estimation of Energy Storage and Its Feasibility Analysis 73

**Figure 15.** Cost of energy (COE) in different cases

combination of RE used as shown in Figure 16.

**Figure 16.** Net present cost (NPC) in different cases

**5.4. Payback period** 

NPC represents present cost of the system. In standalone configuration NPC was very high however storage helped in reducing NPC to an acceptable level by improving the utilization of RE. In Grid connected configuration, storage helped in reducing NPC in every

In this model payback was calculated by comparing one system with another. Payback is the number of years in which the cumulative cash flow switches from negative to positive by comparing storage integrated model with without storage model in grid connected configuration. Cash flow in grid connected PV with storage system compared with grid connected PV base system and it was found that payback period is 4.15 year. Similarly grid

**Net Present Cost (NPC)** 

**Figure 13.** GHG and pollutant gas emission in different cases

**Figure 14.** Renewable fraction (RF) in different cases

### **Cost of Energy (COE)**

COE is the cost of per unit energy in \$/kWh. Stand-alone configuration involving diesel generator was costly therefore COE was very high; however adding storage reduced COE to a reasonable level. In Grid connected configuration in all combination of RE sources, storage reduced the COE and in hybrid system storage reduced COE close to the grid only energy cost as shown in Figure 15.

Estimation of Energy Storage and Its Feasibility Analysis 73

**Figure 15.** Cost of energy (COE) in different cases

#### **Net Present Cost (NPC)**

72 Energy Storage – Technologies and Applications

**Figure 13.** GHG and pollutant gas emission in different cases

**Figure 14.** Renewable fraction (RF) in different cases

COE is the cost of per unit energy in \$/kWh. Stand-alone configuration involving diesel generator was costly therefore COE was very high; however adding storage reduced COE to a reasonable level. In Grid connected configuration in all combination of RE sources, storage reduced the COE and in hybrid system storage reduced COE close to the grid only energy

**Cost of Energy (COE)** 

cost as shown in Figure 15.

NPC represents present cost of the system. In standalone configuration NPC was very high however storage helped in reducing NPC to an acceptable level by improving the utilization of RE. In Grid connected configuration, storage helped in reducing NPC in every combination of RE used as shown in Figure 16.

**Figure 16.** Net present cost (NPC) in different cases

#### **5.4. Payback period**

In this model payback was calculated by comparing one system with another. Payback is the number of years in which the cumulative cash flow switches from negative to positive by comparing storage integrated model with without storage model in grid connected configuration. Cash flow in grid connected PV with storage system compared with grid connected PV base system and it was found that payback period is 4.15 year. Similarly grid

connected wind generator with storage compared with without storage system and found that payback period is 2.67 years. In case of grid connected Hybrid (PV & wind turbine) system with storage compared with same without storage system and found that payback period is 2.05 years. Storage helped in RE utilization that minimizes the use of grid electricity and increased energy sell back to the grid. Therefore it was confirmed that the investment cost of storage integration returns in very short period of time as shown in Figure 17. In Australia solar bonus scheme awards the price of electricity fed into the grid from RE at a rate of \$0.44/kWh [39-40] which is much higher than the utility rate. This ensures that the payback period will be much shorter in Australia.

Estimation of Energy Storage and Its Feasibility Analysis 75

reduced 43.35% of GHG and pollutant gas emission and in all standalone systems it was 100%. In grid connected wind and hybrid system, storage reduced GHG emission more than 100% by selling extra energy to the grid. In grid connected configuration storage improved RF where with PV, wind turbine and hybrid system RF was 59.8%, 98.7%, 98.6% respectively. Storage reduced COE and in grid connected configuration it was as low as presently available grid electricity cost. In hybrid and wind system COE was \$0.316/kWh and \$0.302/kWh respectively. Similarly Storage reduced NPC by 8.3%, 27.13% and 33.95% in grid connected PV, wind and hybrid configurations respectively. Moreover payback time of storage is very short therefore storage integrated RE system is more feasible for

implementation.

**Author details** 

A. B. M. Shawkat Ali

**7. References** 

http://reg.bom.gov.au/.

*Dakota, USA. Wind Energy, 2009.*

*110, D12110, doi:1029/2004JD005462.* 2005. 110.

*Sons Ltd. 2001.*

2004.

Mohammad Taufiqul Arif and Amanullah M. T. Oo

*directly from Ergon Energy. Dated: 05/07/2010.*

*Power Engineering Research Group, Faculty of Sciences, Engineering & Health, Central Queensland University, Bruce Highway, Rockhampton, QLD 4702, Australia* 

*School of Computing Sciences, Faculty of Arts, Business, Informatics & Education, Central Queensland University, Bruce Highway, Rockhampton, QLD 4702, Australia* 

[2] BoM, *Bureau of Meteorology, Australian Government. Available [Online] at:* 

*[Online] at: http://homerenergy.com/.* Homer Help file, 2007.

*energy in Malaysia.* Renewable Energy 36 (2011) 881-888, 2011.

[9] Grigsbay, L.L., *The Electric Power Engineering Handbook.* CRC Press 2001.

[1] ErgonEnergy, *(2010), Capricornia region electric load, Rockhampton load data collected* 

[3] Lambert, T., *How HOMER Calculates the PV Array Power Output, Software available* 

[4] Ahmed M.A. Haidar, P.N.J., Mohd Shawal, *Optimal configuration assessment of renewable* 

[5] Tony Burton, D.S., Nick Jenkins, Ervin Bossanyi, *Wind Energy Handbook, John Wiley &* 

[6] Aaron Knoll, K.K., *Residential and Commercial scale distributed wind energy in North* 

[7] Masters, G.M., *Renewable and Efficient Electric Power Systems.* John Wiley & Sons, Inc.,

[8] L. Cristina, A.a.J.Z.M., *Evaluation of global wind power, Journal of Geophysical Research, Vol-*

**Figure 17.** Payback period of storage in three different cases

## **6. Conclusion**

Storage integrated RE system was analyzed for a residential load in Rockhampton, Australia. Estimation of required storage for the RE system was calculated. Estimation steps were developed and estimation of required storage was done in grid connected PV, wind and hybrid systems. It was found that to support daily load of 15.7kWh/day in grid connected PV system minimum 665.90Ah of storage, in grid connected wind turbine minimum 582.356Ah of storage and in hybrid system minimum 543.62Ah of storage required at 24V DC system voltage.

Model was developed for feasibility analysis of storage with RE. Model was analyzed in standalone and grid connected configurations. Analysis was conducted to observe the storage influences over the GHG emission, RF, COE and NPC indexes. It was found by analyzing the output data from the optimized model that storage has great influence on improving RE utilization.

It was evident from the analysis that storage helped significantly in reducing GHG & other pollutant gas emission, reduced COE, improved RF and reduced NPC. Comparing without and with storage system model, it was found that in grid connected PV system, storage reduced 43.35% of GHG and pollutant gas emission and in all standalone systems it was 100%. In grid connected wind and hybrid system, storage reduced GHG emission more than 100% by selling extra energy to the grid. In grid connected configuration storage improved RF where with PV, wind turbine and hybrid system RF was 59.8%, 98.7%, 98.6% respectively. Storage reduced COE and in grid connected configuration it was as low as presently available grid electricity cost. In hybrid and wind system COE was \$0.316/kWh and \$0.302/kWh respectively. Similarly Storage reduced NPC by 8.3%, 27.13% and 33.95% in grid connected PV, wind and hybrid configurations respectively. Moreover payback time of storage is very short therefore storage integrated RE system is more feasible for implementation.

## **Author details**

74 Energy Storage – Technologies and Applications

connected wind generator with storage compared with without storage system and found that payback period is 2.67 years. In case of grid connected Hybrid (PV & wind turbine) system with storage compared with same without storage system and found that payback period is 2.05 years. Storage helped in RE utilization that minimizes the use of grid electricity and increased energy sell back to the grid. Therefore it was confirmed that the investment cost of storage integration returns in very short period of time as shown in Figure 17. In Australia solar bonus scheme awards the price of electricity fed into the grid from RE at a rate of \$0.44/kWh [39-40] which is much higher than the utility rate. This

Storage integrated RE system was analyzed for a residential load in Rockhampton, Australia. Estimation of required storage for the RE system was calculated. Estimation steps were developed and estimation of required storage was done in grid connected PV, wind and hybrid systems. It was found that to support daily load of 15.7kWh/day in grid connected PV system minimum 665.90Ah of storage, in grid connected wind turbine minimum 582.356Ah of storage and in hybrid system minimum 543.62Ah of storage

Model was developed for feasibility analysis of storage with RE. Model was analyzed in standalone and grid connected configurations. Analysis was conducted to observe the storage influences over the GHG emission, RF, COE and NPC indexes. It was found by analyzing the output data from the optimized model that storage has great influence on

It was evident from the analysis that storage helped significantly in reducing GHG & other pollutant gas emission, reduced COE, improved RF and reduced NPC. Comparing without and with storage system model, it was found that in grid connected PV system, storage

ensures that the payback period will be much shorter in Australia.

**Figure 17.** Payback period of storage in three different cases

**6. Conclusion** 

required at 24V DC system voltage.

improving RE utilization.

Mohammad Taufiqul Arif and Amanullah M. T. Oo *Power Engineering Research Group, Faculty of Sciences, Engineering & Health, Central Queensland University, Bruce Highway, Rockhampton, QLD 4702, Australia* 

#### A. B. M. Shawkat Ali

*School of Computing Sciences, Faculty of Arts, Business, Informatics & Education, Central Queensland University, Bruce Highway, Rockhampton, QLD 4702, Australia* 

## **7. References**


[10] Paul Denholm, E.E., Brendan Kirby, and Michael Milligan, *The Role of Energy Storage with Renewable Electricity Generation.* Technical Report, NREL/TP-6A2-47187, January 2010.

Estimation of Energy Storage and Its Feasibility Analysis 77

[26] Celik, *A.N., Techno-economic analysis of autonomous PV-Wind hybrid energy systems using different sizing methods. Energy Conversion and Management, 2003. 44(12): P. 1951-1968.* [27] Barley, *DC, Winn BC. Optimal dispatch strategy in remote hybrid power systems, Solar* 

[28] BoM, *Bureau of Meteorology (2011), Australian Government. Available [Online] at:* 

[29] ErgonEnergy, *Understanding your Electricity Bill, Information about average consumption, Available [Online] at: http://www.ergon.com.au/your-business/accounts--and--*

*[30]* PV-Price, *Sun Solar System, Available [Online] at: www.sunsolarsystem.com.au (access date:* 

http://www.goodhewsolar.com.au/customPages/goodhew-electrical-%26-solar-offershomeowners-the-most-affordable-quality-solar-systems-on-the-market.?subSiteId=1

[32] Inverter-Cost, *SMA Sunny Boy Grid tie Inverter 7000W SB7000US price, Available [Online]* 

1700?ver=gg&gclid=CMO0gIOzna4CFYVMpgod7T1OHg (access date: 16/03/2012). [34] WindPower, B., *Wind Turbine Retail Price list, Available [Online[ at: http://productionimages.webapeel.com/bergey/assets/2012/3/6/98837/PriceList-March.pdf Access date:* 

[35] Winturbine\_Price, *Ecodirect, Clean Energy Solution, Windturbine Price, Available [Online] at: http://www.ecodirect.com/Bergey-Windpower-BWC-10kW-p/bergey-windpower-bwc-10kw-*

http://www.ergon.com.au/your-business/accounts--and--billing/electricity-prices. [37] ALTE-Store, *Battery Price, Trojan T-105 6V, 225AH (20HR) Flooded Lead Acid Battery, Available [Online] at: http://www.altestore.com/store/Deep-Cycle-Batteries/Batteries-Flooded-*

[38] RedEnergy, *Pricing definition for Electricity customers, NSW, Available [Online] at:* 

http://futuresustainability.rtrk.com.au/?scid=80507&kw=4858156&pub\_cr\_id=171647038

*Lead-Acid/Trojan-T-105-6V-225AH-20HR-Flooded-Lead-Acid-Battery/p1771/.*

[39] Future\_Sustainability, *Rebates for Solar Power, Available [Online] at:* 

*http://www.redenergy.com.au/docs/NSW-Pricing-DEFINITIONS-0311.pdf.* 2011.

*http://www.google.com/products/catalog?hl=en&q=sunny+boy+grid+tie+inverter+price&gs\_sm= 3&gs\_upl=2378l7840l1l8534l11l10l0l1l1l0l223l2055l0.3.7l11l0&bav=on.2,or.r\_gc.r\_pw.,cf.osb*

*Energy 1996;58:1 65-79.*

(access date: 15/03/2012).

*&biw=1680&bih=831&um=1&ie=UTF-*

*15/03/2012).* 

*at:* 

*09/04/2012.*

77.

*http://reg.bom.gov.au/ Data collected on 03/06/2011.*

*billing/understanding-your-bill access date: 10/04/2012.*

[31] PV-Price, *Goodhew Electrical and Solar, Available [Online] at:* 

*8&tbm=shop&cid=10871923140935237408&sa=X&ei=z0g6T-*

[33] Inverter-Cost, *SMA Sunny Boy 1700 Price, Available [Online] at:*  http://www.solarmatrix.com.au/special-offers/sunny-boy-

*ex.htm Access date: 29/02/2012.* Online Information. [36] ErgonEnergy, *Electricity tariffs and prices, available [Online] at:* 

*T0LayziQfnkZGQCg&ved=0CGkQ8wIwAQ (access date: 16/03/2012).*


*http://www.google.com/products/catalog?hl=en&q=sunny+boy+grid+tie+inverter+price&gs\_sm= 3&gs\_upl=2378l7840l1l8534l11l10l0l1l1l0l223l2055l0.3.7l11l0&bav=on.2,or.r\_gc.r\_pw.,cf.osb &biw=1680&bih=831&um=1&ie=UTF-*

*8&tbm=shop&cid=10871923140935237408&sa=X&ei=z0g6T-*

76 Energy Storage – Technologies and Applications

Access date 24 November 2011.

2010.

p. 90-95.

1013™-2007.

24.

*1490.*

Release, July 2008, 2008.

*(PV) Systems, Available [Online] at:* 

77 (access date: 20/03/2012).

https://analysis.nrel.gov/homer/.

[10] Paul Denholm, E.E., Brendan Kirby, and Michael Milligan, *The Role of Energy Storage with Renewable Electricity Generation.* Technical Report, NREL/TP-6A2-47187, January

[11] Wessells, C., Stanford University News*, Nanoparticle electrode for batteries could make* 

http://news.stanford.edu/news/2011/november/longlife-power-storage-112311.html

[13] Jozef, *Paska et. al, Technical and Economic Aspects of Electricity Storage Systems Co-operating* 

[14] Mohammad T Arif, A.M.T.O., A B M Shawkat Ali, Md. Fakhrul Islam, *Significance of Storage and feasibility analysis of Renewable energy with storage system.* Proceedings of the IASTED International Conference on Power and Energy Systems (Asia PES 2010), 2010:

[15] Dan T. Ton, C.J.H., Georgianne H. Peek, and John D. Boyes, *Solar Energy Grid Integration Systems –Energy Storage (SEGIS-ES).* SANDIA REPORT, SAND2008-4247, Unlimited

[16] Goel, G.B.S.a.L., *A study on optimal sizing of stand-alone photovoltaic stations.* IEEE

[17] Nabiha BRAHMI, S.S., Maher CHAABENE, *Sizing of a mini autonomous hybrid electric* 

[18] IEEE, *IEEE Recommended Practice for Sizing Lead-Acid Batteries for Stand-Alone Photovoltaic* 

[19] IEEE, *IEEE Std 1561-2007, IEEE Guide for Optimizing the Performance and Life of Lead-Acid* 

[20] IEEE, *IEEE Standard for Interconnecting Distributed Resources With Electric Power Systems,* 

[21] Renewable Energy, T.I.p.o.T., *Estimating PV System Size and Cost.* SECO Fact Sheet no.

http://futuresustainability.rtrk.com.au/?scid=80507&kw=4858156&pub\_cr\_id=171647038

[23] Matters, E., *SMA Sunny Boy 3800W Grid-connected Inverter, Available [Online] at: http://www.energymatters.com.au/sma-sunny-boy-3800watt-grid-connect-inverter-p-412.html.* [24] G.J., *Dalton, Lockington, D. A. & Baldock, T. E. (2008) Feasibility analysis of stand-alone renewable energy supply options for a large hotel. Renewable Energy, vol 33, issue 7, P-1475-*

http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4280849. IEEE Standard

Transactions on Energy Conversion, Vol. 13, No. 4, December 1998, 1998.

*large-scale power storage on the energy grid feasible,* Available [Online] at:

*with Renewable Energy Sources. .* 10th Conference EPQU 2009.

*grid.* International Renewable Energy Congress, 2009.

*Batteries in Remote Hybrid Power Systems.* Standard, 2007.

*in IEEE Std 1547-2003. p. 0\_1-16.* Standard, 2003.

[22] PV, L., *LG Polycrystalline PV Module, Available [Online] at:* 

[25] HOMER, *Analysis of micro powersystem options, Available [Online] at:* 

[12] Ter-Gazarian, A., *Energy Storage for Power Systems. .* Peter Peregrinus Ltd, 1994.

*T0LayziQfnkZGQCg&ved=0CGkQ8wIwAQ (access date: 16/03/2012).*

	- [40] Depertment of Employment, E.D.I., Queensland, *Solar Bonus Scheme, Queensland Government, Office of Clean Energy, Available [Online] at:*  http://www.cleanenergy.qld.gov.au/demand-side/solar-bonusscheme.htm?utm\_source=WWW2BUSINESS&utm\_medium=301&utm\_campaign= redirection.

**Chapter 3** 

© 2013 Krivik and Baca, licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

© 2013 Krivik and Baca, licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

**Electrochemical Energy Storage** 

Electrochemical energy storage covers all types of secondary batteries. Batteries convert the chemical energy contained in its active materials into electric energy by an electrochemical

At present batteries are produced in many sizes for wide spectrum of applications. Supplied powers move from W to the hundreds of kW (compare battery for power supply of pace

Common commercially accessible secondary batteries according to used electrochemical

Standard batteries (lead acid, Ni-Cd) modern batteries (Ni-MH, Li–ion, Li-pol), special batteries (Ag-Zn, Ni-H2), flow batteries (Br2-Zn, vanadium redox) and high temperature

Lead acid battery when compared to another electrochemical source has many advantages. It is low price and availability of lead, good reliability, high voltage of cell (2 V), high electrochemical effectivity, cycle life is from several hundreds to thousands of cycles. Thanks to these characteristics is now the most widely used secondary electrochemical source of electric energy and represent about 60% of installed power from all types of secondary batteries. Its disadvantage is especially weight of lead and consequently lower specific

Lead-acid batteries are suitable for medium and large energy storage applications because

they offer a good combination of power parameters and a low price.

makers and battery for heavy motor vehicle or for power station).

system can be divided to the following basic groups:

Additional information is available at the end of the chapter

Petr Krivik and Petr Baca

http://dx.doi.org/10.5772/52222

oxidation-reduction reverse reaction.

batteries (Na-S, Na–metalchloride).

energy in the range 30-50 Wh/kg.

**2. Standard batteries** 

**2.1. Lead acid battery** 

**1. Introduction** 

**Chapter 3** 

## **Electrochemical Energy Storage**

Petr Krivik and Petr Baca

78 Energy Storage – Technologies and Applications

redirection.

[40] Depertment of Employment, E.D.I., Queensland, *Solar Bonus Scheme, Queensland* 

scheme.htm?utm\_source=WWW2BUSINESS&utm\_medium=301&utm\_campaign=

*Government, Office of Clean Energy, Available [Online] at:* 

http://www.cleanenergy.qld.gov.au/demand-side/solar-bonus-

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/52222

**1. Introduction** 

Electrochemical energy storage covers all types of secondary batteries. Batteries convert the chemical energy contained in its active materials into electric energy by an electrochemical oxidation-reduction reverse reaction.

At present batteries are produced in many sizes for wide spectrum of applications. Supplied powers move from W to the hundreds of kW (compare battery for power supply of pace makers and battery for heavy motor vehicle or for power station).

Common commercially accessible secondary batteries according to used electrochemical system can be divided to the following basic groups:

Standard batteries (lead acid, Ni-Cd) modern batteries (Ni-MH, Li–ion, Li-pol), special batteries (Ag-Zn, Ni-H2), flow batteries (Br2-Zn, vanadium redox) and high temperature batteries (Na-S, Na–metalchloride).

## **2. Standard batteries**

## **2.1. Lead acid battery**

Lead acid battery when compared to another electrochemical source has many advantages. It is low price and availability of lead, good reliability, high voltage of cell (2 V), high electrochemical effectivity, cycle life is from several hundreds to thousands of cycles. Thanks to these characteristics is now the most widely used secondary electrochemical source of electric energy and represent about 60% of installed power from all types of secondary batteries. Its disadvantage is especially weight of lead and consequently lower specific energy in the range 30-50 Wh/kg.

Lead-acid batteries are suitable for medium and large energy storage applications because they offer a good combination of power parameters and a low price.

© 2013 Krivik and Baca, licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2013 Krivik and Baca, licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

## *2.1.1. Battery composition and construction*

Construction of lead acid (LA) battery depends on usage. It is usually composed of some series connected cells. Main parts of lead acid battery are electrodes, separators, electrolyte, vessel with lid, ventilation and some other elements.

Electrochemical Energy Storage 81

made from lead alloys (pure lead would be too soft); it is used Pb-Ca or Pb-Sb alloys, with mixture of additives as Sn, Cd and Se, that improve corrosion resistance and make higher

Active material is made from lead oxide PbO pasted onto a grid and then electrochemically converted into reddish brown lead dioxide PbO2 on positive electrode and on grey spongy

Separators electrically separate positive electrode from negative. They have four functions:

2. to act as a mechanical spacer which holds the plates in the prescribed position,

4. to permit both the free diffusion of electrolyte and the migration of ions.

rubber, polyvinyl chloride (PVC), polyethylene (PE), and glass-microfibre.

3. to help retain the active materials in close contact with the grid,

1. to provide electrical insulation between positive and negative plate and to prevent short

The materials used for separators can be wood veneers, cellulose (paper), usually stiffened with a phenol-formaldehyde resin binder, and those made from synthetic materials, e.g.,

Electrolyte is aqueous solution of H2SO4 with density of 1.22-1.28 g/cm3. Mostly it is liquid, covered battery plates. Sometimes it is transformed to the form of gel, or completely

Vessel must to withstand straining caused by weight of inner parts of battery and inner pressure from gas rising during cycling. The most used material is polypropylene, but also, PVC, rubber etc. If overpressure rises inside classical battery during charging, problem is

There are some major types of battery construction: prismatic construction with grid or tubular plates, cylindrical construction (spiral wound or disc plates) or bipolar construction.

PbO2 + Pb + 2H2SO4 → 2PbSO4 + 2H2O E0 = +2.048 V (1)

According to the usage and construction, lead acid batteries split into stationary, traction

Stationary battery ensures uninterrupted electric power supply in case of failure in distributing network. During its service life battery undergo only few cycles. Battery life is

mechanical strength.

circuits,

absorbed in separators.

solved by valve placed mostly in lid.

Overall chemical reaction during discharge is:

Reaction proceeds in opposite direction during charge.

*2.1.2. Principle of operation* 

*2.1.3. Types of LA batteries* 

and automotive batteries.

as many as 20 years.

lead Pb on negative electrode.

**Figure 1.** Scheme of prismatic and spiral wound construction of LA battery

Electrode consists of grid and of active mass. Grid as bearing structure of electrode must be mechanically proof and positive electrode grid must be corrosion proof. Corrosion converts lead alloy to lead oxides with lower mechanical strength and electric conductivity. Grids are made from lead alloys (pure lead would be too soft); it is used Pb-Ca or Pb-Sb alloys, with mixture of additives as Sn, Cd and Se, that improve corrosion resistance and make higher mechanical strength.

Active material is made from lead oxide PbO pasted onto a grid and then electrochemically converted into reddish brown lead dioxide PbO2 on positive electrode and on grey spongy lead Pb on negative electrode.

Separators electrically separate positive electrode from negative. They have four functions:


The materials used for separators can be wood veneers, cellulose (paper), usually stiffened with a phenol-formaldehyde resin binder, and those made from synthetic materials, e.g., rubber, polyvinyl chloride (PVC), polyethylene (PE), and glass-microfibre.

Electrolyte is aqueous solution of H2SO4 with density of 1.22-1.28 g/cm3. Mostly it is liquid, covered battery plates. Sometimes it is transformed to the form of gel, or completely absorbed in separators.

Vessel must to withstand straining caused by weight of inner parts of battery and inner pressure from gas rising during cycling. The most used material is polypropylene, but also, PVC, rubber etc. If overpressure rises inside classical battery during charging, problem is solved by valve placed mostly in lid.

There are some major types of battery construction: prismatic construction with grid or tubular plates, cylindrical construction (spiral wound or disc plates) or bipolar construction.

## *2.1.2. Principle of operation*

80 Energy Storage – Technologies and Applications

*2.1.1. Battery composition and construction* 

vessel with lid, ventilation and some other elements.

**Figure 1.** Scheme of prismatic and spiral wound construction of LA battery

Electrode consists of grid and of active mass. Grid as bearing structure of electrode must be mechanically proof and positive electrode grid must be corrosion proof. Corrosion converts lead alloy to lead oxides with lower mechanical strength and electric conductivity. Grids are

Construction of lead acid (LA) battery depends on usage. It is usually composed of some series connected cells. Main parts of lead acid battery are electrodes, separators, electrolyte,

Overall chemical reaction during discharge is:

$$\text{PbO} + \text{Pb} + 2\text{H} \text{SiO} \downarrow \xrightarrow{} 2\text{PbSO} + 2\text{H} \text{O} \qquad \qquad \text{E}^0 = +2.048 \text{ V} \tag{1}$$

Reaction proceeds in opposite direction during charge.

## *2.1.3. Types of LA batteries*

According to the usage and construction, lead acid batteries split into stationary, traction and automotive batteries.

Stationary battery ensures uninterrupted electric power supply in case of failure in distributing network. During its service life battery undergo only few cycles. Battery life is as many as 20 years.

Traction battery is used for power supply of industrial trucks, delivery vehicles, electromobiles, etc. It works in cyclic regime of deep charge–discharge. Cycle life of the battery is about 5 years (1000 of charge–discharge cycles).

Automotive battery is used for cranking automobile internal combustion engines and also for supporting devices which require electrical energy when the engine is not running. It must be able of supplying short but intense discharge current. It is charged during running of engine.

According to the maintenance operation lead acid batteries could be branched into conventional batteries (i.e., those with free electrolyte, so-called 'flooded' designs), requiring regular maintenance and valve-regulated lead-acid (VRLA) maintenance free batteries.

## *2.1.4. VRLA batteries*

Originally, the battery worked with its plates immersed in a liquid electrolyte and the hydrogen and the oxygen produced during overcharge were released into the atmosphere. The lost gases reflect a loss of water from the electrolyte and it had to be filled in during maintenance operation. Problems with water replenishing were overcome by invention of VRLA (valve regulated lead acid) batteries.

The VRLA battery is designed to operate with help of an internal oxygen cycle, see Fig. 2. Oxygen liberated during the latter stages of charging, and during overcharging, on the positive electrode, i.e.

$$\text{HO} \xrightarrow{\quad} \text{2H} + 1/2\text{O} + 2\text{e} \quad \text{ (1a)}$$

Electrochemical Energy Storage 83

**Figure 2.** Internal oxygen cycle in a valve regulated lead acid cell (Nelson, 2001).

negative electrode. Gas release from the cell then falls rapidly (Rand et al., 2004).

1. positive plate expansion and positive active mass fractioning, 2. water loss brought about by gassing or by a high temperature,

4. incomplete charging causing active mass sulphation,

Lead acid batteries can be affected by one or more of the following failure mechanisms:

6. negative active mass sulphation (batteries in partial state of charge (PSoC) cycling batteries in hybrid electric vehicles (HEV) and batteries for remote area power supply

Repetitive discharge and charge of the LA battery causes expansion of the positive active mass because product of the discharge reaction PbSO4 occupies a greater volume than the positive active material PbO2. Charging of the cell restores most of the lead dioxide, but not within the original volume. The negative active mass does not show the same tendency to expand. Reason could be that lead is softer than lead dioxide and that is why the negative active material is more compressed during discharge as the conversion from lead to the more voluminous lead sulphate proceeds. Another reason could be that spongy lead

*2.1.5. Failure mechanisms of LA batteries* 

3. acid stratification,

5. positive grid corrosion,

(RAPS) applications).

When the cell is filled with electrolyte, the oxygen cycle is impossible because oxygen diffuses through the electrolyte very slow. On the end of charge, first oxygen (from the positive), and then both oxygen (from the positive) and hydrogen (from the negative), are liberated and they are released through the pressure valve. Gassing causes loss of water and opens gas spaces due to drying out of the gel electrolyte or a liquid electrolyte volume decrease in the AGM separators). It allows the transfer of oxygen from the positive to the

travels through a gas space in separator to the negative electrode where is reduced to the water:

$$\text{Pb} + 1/2\text{Oh} + \text{HzSO4} \rightarrow \text{PbSO4} + \text{HzO} + \text{Heat} \tag{1b}$$

The oxygen cycle, defined by reactions (1a) and (1b), moves the potential of the negative electrode to a less negative value and, consequently, the rate of hydrogen evolution decreases. The small amount of hydrogen that could be produced during charging is released by pressure valve. The produced lead sulphate is immediately reduced to lead via the reaction (1c), because the plate is simultaneously on charge reaction:

$$\text{PbSO} + 2\text{H}^+ + 2\text{e}^- \rightleftharpoons \text{Pb} + \text{HSO} \tag{1c}$$

The sum of reactions (1a), (1b) and (1c) is zero. Part of the electrical energy delivered to the cell is consumed by the internal oxygen recombination cycle and it is converted into heat.

There are two designs of VRLA cells which provide the internal oxygen cycle. One has the electrolyte immobilized as a gel (gel batteries), the other has the electrolyte held in an AGM separator (AGM batteries). Gas can pass through crack in the gel, or through channels in the AGM separator.

**Figure 2.** Internal oxygen cycle in a valve regulated lead acid cell (Nelson, 2001).

When the cell is filled with electrolyte, the oxygen cycle is impossible because oxygen diffuses through the electrolyte very slow. On the end of charge, first oxygen (from the positive), and then both oxygen (from the positive) and hydrogen (from the negative), are liberated and they are released through the pressure valve. Gassing causes loss of water and opens gas spaces due to drying out of the gel electrolyte or a liquid electrolyte volume decrease in the AGM separators). It allows the transfer of oxygen from the positive to the negative electrode. Gas release from the cell then falls rapidly (Rand et al., 2004).

## *2.1.5. Failure mechanisms of LA batteries*

Lead acid batteries can be affected by one or more of the following failure mechanisms:


82 Energy Storage – Technologies and Applications

of engine.

*2.1.4. VRLA batteries* 

positive electrode, i.e.

water:

AGM separator.

VRLA (valve regulated lead acid) batteries.

H2O → 2H+ + 1/2O2 + 2e-

PbSO4 + 2H+ + 2e-

the reaction (1c), because the plate is simultaneously on charge reaction:

battery is about 5 years (1000 of charge–discharge cycles).

Traction battery is used for power supply of industrial trucks, delivery vehicles, electromobiles, etc. It works in cyclic regime of deep charge–discharge. Cycle life of the

Automotive battery is used for cranking automobile internal combustion engines and also for supporting devices which require electrical energy when the engine is not running. It must be able of supplying short but intense discharge current. It is charged during running

According to the maintenance operation lead acid batteries could be branched into conventional batteries (i.e., those with free electrolyte, so-called 'flooded' designs), requiring regular maintenance and valve-regulated lead-acid (VRLA) maintenance free batteries.

Originally, the battery worked with its plates immersed in a liquid electrolyte and the hydrogen and the oxygen produced during overcharge were released into the atmosphere. The lost gases reflect a loss of water from the electrolyte and it had to be filled in during maintenance operation. Problems with water replenishing were overcome by invention of

The VRLA battery is designed to operate with help of an internal oxygen cycle, see Fig. 2. Oxygen liberated during the latter stages of charging, and during overcharging, on the

travels through a gas space in separator to the negative electrode where is reduced to the

The oxygen cycle, defined by reactions (1a) and (1b), moves the potential of the negative electrode to a less negative value and, consequently, the rate of hydrogen evolution decreases. The small amount of hydrogen that could be produced during charging is released by pressure valve. The produced lead sulphate is immediately reduced to lead via

The sum of reactions (1a), (1b) and (1c) is zero. Part of the electrical energy delivered to the cell is consumed by the internal oxygen recombination cycle and it is converted into heat.

There are two designs of VRLA cells which provide the internal oxygen cycle. One has the electrolyte immobilized as a gel (gel batteries), the other has the electrolyte held in an AGM separator (AGM batteries). Gas can pass through crack in the gel, or through channels in the

Pb + 1/2O2 + H2SO4 → PbSO4 + H2O + Heat (1b)

, (1a)

→ Pb + H2SO4 (1c)


Repetitive discharge and charge of the LA battery causes expansion of the positive active mass because product of the discharge reaction PbSO4 occupies a greater volume than the positive active material PbO2. Charging of the cell restores most of the lead dioxide, but not within the original volume. The negative active mass does not show the same tendency to expand. Reason could be that lead is softer than lead dioxide and that is why the negative active material is more compressed during discharge as the conversion from lead to the more voluminous lead sulphate proceeds. Another reason could be that spongy lead

contains bigger pores than pores in lead dioxide and therefore is more easily able to absorb a lead sulphate without expansion of a negative active mass. Progressive expansion of the positive electrode causes an increasing fraction of the positive active material. This material becomes to be electrically disconnected from the current collection process and it causes decreasing of the cell capacity. (Calabek et al., 2001).

Electrochemical Energy Storage 85

grow to big ones. Big crystals of lead sulphate increase internal resistance of the cell and

during charging it is hardly possible to convert them back to the active mass.

**Figure 4.** SEM images of negative active mass. Sulphation on the left, healthy state on the right

to collapse of the positive electrode.

**Figure 5.** Positive grid corrosion

**2.2. Ni-Cd battery** 

During charge the positive grid is subject to corrosion. Lead collector turns on lead dioxide or lead sulphate. The rate of this process depends on the grid composition and microstructure, also on plate potential, electrolyte composition and temperature of the cell. The corrosion products have usually a bigger electric resistance than positive grid. In extreme cases, corrosion could result to disintegration of the positive grid and consequently

They are main representative of batteries with positive Nickel electrode; other possible

systems could be system Ni-Fe and Ni–Zn, Ni-H2 or Ni–MH.

**Figure 3.** Positive active mass fractioning

Gas evolving during overcharge leads to reduction of the volume of the electrolyte. Some of the active material consequently loses contact with the electrodes. Drying out increases the internal resistance of the battery which causes excessive rise of temperature during charging and this process accelerates water loss through evaporation.

During charge, sulphuric acid is produced between the electrodes and there is a tendency for acid of higher concentration, which has a greater relative density, to fall to the bottom of the lead acid cell. Acid stratification can be caused also by preferential discharge of upper parts of the cell, because of lower ohmic resistance of these parts. Concentration of electrolyte in the upper part of the cell is temporarily lower than on the bottom of the cell. It leads to discharge of the bottom parts and charge of the upper parts of the cell. The vertical concentration gradient of sulphuric acid can give rise to non uniform utilization of active mass and, consequently, shortened service life through the irreversible formation of PbSO4 (Ruetschi, 2004).

When the electrodes are repeatedly not fully charged, either because of a wrong charging procedure or as a result of physical changes that keep the electrode from reaching an adequate potential (antimony poisoning of negative electrode), then a rapid decreasing in battery capacity may occur because of progressive accumulation of lead sulphate in active mass. Sulphation is creation of insulation layer of lead sulphate on the electrode surface. It leads to inhibition of electrolyte contact with active mass. Sulphation grows during the long term standing of the battery in discharge state, in case of electrolyte stratification, or incomplete charging. In the course of sulphation originally small crystals of lead sulphate grow to big ones. Big crystals of lead sulphate increase internal resistance of the cell and during charging it is hardly possible to convert them back to the active mass.

**Figure 4.** SEM images of negative active mass. Sulphation on the left, healthy state on the right

During charge the positive grid is subject to corrosion. Lead collector turns on lead dioxide or lead sulphate. The rate of this process depends on the grid composition and microstructure, also on plate potential, electrolyte composition and temperature of the cell. The corrosion products have usually a bigger electric resistance than positive grid. In extreme cases, corrosion could result to disintegration of the positive grid and consequently to collapse of the positive electrode.

**Figure 5.** Positive grid corrosion

## **2.2. Ni-Cd battery**

84 Energy Storage – Technologies and Applications

**Figure 3.** Positive active mass fractioning

(Ruetschi, 2004).

and this process accelerates water loss through evaporation.

decreasing of the cell capacity. (Calabek et al., 2001).

contains bigger pores than pores in lead dioxide and therefore is more easily able to absorb a lead sulphate without expansion of a negative active mass. Progressive expansion of the positive electrode causes an increasing fraction of the positive active material. This material becomes to be electrically disconnected from the current collection process and it causes

Gas evolving during overcharge leads to reduction of the volume of the electrolyte. Some of the active material consequently loses contact with the electrodes. Drying out increases the internal resistance of the battery which causes excessive rise of temperature during charging

During charge, sulphuric acid is produced between the electrodes and there is a tendency for acid of higher concentration, which has a greater relative density, to fall to the bottom of the lead acid cell. Acid stratification can be caused also by preferential discharge of upper parts of the cell, because of lower ohmic resistance of these parts. Concentration of electrolyte in the upper part of the cell is temporarily lower than on the bottom of the cell. It leads to discharge of the bottom parts and charge of the upper parts of the cell. The vertical concentration gradient of sulphuric acid can give rise to non uniform utilization of active mass and, consequently, shortened service life through the irreversible formation of PbSO4

When the electrodes are repeatedly not fully charged, either because of a wrong charging procedure or as a result of physical changes that keep the electrode from reaching an adequate potential (antimony poisoning of negative electrode), then a rapid decreasing in battery capacity may occur because of progressive accumulation of lead sulphate in active mass. Sulphation is creation of insulation layer of lead sulphate on the electrode surface. It leads to inhibition of electrolyte contact with active mass. Sulphation grows during the long term standing of the battery in discharge state, in case of electrolyte stratification, or incomplete charging. In the course of sulphation originally small crystals of lead sulphate

They are main representative of batteries with positive Nickel electrode; other possible systems could be system Ni-Fe and Ni–Zn, Ni-H2 or Ni–MH.

## *2.2.1. Battery composition and construction*

The nickel cadmium cell has positive electrode from nickel hydroxide and negative electrode from metallic cadmium, an electrolyte is potassium hydroxide. The nickel cadmium battery is produced in a wide range of commercially important battery systems from sealed maintenance free cells (capacities of 10 mAh - 20 Ah) to vented standby power units (capacities of 1000 Ah and more). Nickel cadmium battery has long cycle life, overcharge capability, high rates of discharge and charge, almost constant discharge voltage and possibility of operation at low temperature. But, the cost of cadmium is several times that of lead and the cost of nickel cadmium cell construction is more expensive than that of lead acid cell. And there is also problem with the manipulation of toxic cadmium. But also low maintenance and good reliability have made it an ideal for a number of applications such (emergency lighting, engine starting, portable television receivers, hedge trimmers, electric shavers, aircraft and space satellite power systems).

Electrochemical Energy Storage 87

In addition, carbon dioxide in the air can react with KOH in the electrolyte to form K2CO3, and CdCO3 can be formed on the negative plates. Both of these compounds increase the

Ni-Cd batteries suffer from the memory effect (see also chapter Ni-MH battery). Besides Ni-

The sealed nickel metal hydride cell has with hydrogen absorbed in a metal alloy as the active negative material. When compare with Ni-Cd cell it is not only increases the energy density, but also it is a more environmentally friendly power source. The nickel metal hydride cell, however, has high selfdischarge and is less tolerant to overcharge than the Ni-

Positive electrode is NiOOH, negative electrode contains hydrogen absorption alloys. They can absorb over a thousand times their own volume of hydrogen: Alloys usually consist of two metals. First absorbs hydrogen exothermically, a second endothermically. They serve as a catalyst for the dissociative adsorption of atomic hydrogen into the alloy lattice. Examples of used metals: Pd, V, Ti, Zr, Ni, Cr, Co, Sn, Fe, lanthanides and others. The AB2 series

Design of the cylindrical and prismatic sealed Ni-MH cells are similar as with a nickel cadmium cells (see Fig. 7). Hydrophilic polypropylene separator is used in Ni-MH cell.

internal resistance and lower the capacity of the Ni-Cd batteries.

Cd batteries also suffer from high rate of self-discharge at high temperatures.

**Figure 6.** Scheme of spiral wound and prismatic construction of Ni-Cd battery

**3. Modern batteries** 

*3.1.1. Battery composition and construction* 

(ZrNi2) and the AB5 series (LaNi5) are usually used.

**3.1. Ni-MH battery** 

Cd cell.

Depending on construction, nickel cadmium cells have energy densities in the range 40- 60 Wh/kg (50-150 Wh/dm3). Cycle life is moving from several hundreds for sealed cells to several thousands for vented cells.

Cell construction is branched to two types. First using pocket plate electrodes (in vented cells). The active material is found in pockets of finely perforated nickel plated sheet steel. Positive and negative plates are then separated by plastic pins or ladders and plate edge insulators. Second using sintered, bonded or fibre plate electrodes (in both vented and sealed cells). In sintered plate electrodes, a porous sintered nickel electrode is sintered in belt furnace in reducing atmosphere at 800 to 1000°C. Active material is distributed within the pores. In sintered plate cells, a special woven or felted nylon separator is used. It permits oxygen diffusion (oxygen cycle) in sealed cells. In the most common version, a spiral or prismatic construction of cells is used.

The electrolyte is an aqueous solution KOH (concentration of 20-28% by weight and a density of 1.18-1.27 g/cm3 at 25°C). 1-2% of LiOH is usually added to electrolyte to minimize coagulation of the NiOOH electrode during charge/discharge cycling. For low temperature applications, the more concentrated KOH solution is used. When it is operating at high temperature it is sometimes used aqueous NaOH electrolyte.

### *2.2.2. Principle of operation*

The overall cell reaction during discharge:

$$2\text{NiOOH} + \text{Cd} + 2\text{H}\_2\text{O} \xrightarrow{\rightharpoonup} 2\text{Ni(OH)} + \text{Cd(OH)}\text{:}\qquad \text{E}^0 = +1.30\text{ V}\tag{2}$$

It is notable that amount of water in the electrolyte falls during discharge. Ni-Cd batteries are designed as positive limited utilizing oxygen cycle. The oxygen evolved at the positive electrode during charge difuses to the negative electrode and reacts with cadmium to form Cd(OH)2.

In addition, carbon dioxide in the air can react with KOH in the electrolyte to form K2CO3, and CdCO3 can be formed on the negative plates. Both of these compounds increase the internal resistance and lower the capacity of the Ni-Cd batteries.

Ni-Cd batteries suffer from the memory effect (see also chapter Ni-MH battery). Besides Ni-Cd batteries also suffer from high rate of self-discharge at high temperatures.

## **3. Modern batteries**

## **3.1. Ni-MH battery**

86 Energy Storage – Technologies and Applications

several thousands for vented cells.

prismatic construction of cells is used.

The overall cell reaction during discharge:

*2.2.2. Principle of operation* 

Cd(OH)2.

*2.2.1. Battery composition and construction* 

electric shavers, aircraft and space satellite power systems).

temperature it is sometimes used aqueous NaOH electrolyte.

The nickel cadmium cell has positive electrode from nickel hydroxide and negative electrode from metallic cadmium, an electrolyte is potassium hydroxide. The nickel cadmium battery is produced in a wide range of commercially important battery systems from sealed maintenance free cells (capacities of 10 mAh - 20 Ah) to vented standby power units (capacities of 1000 Ah and more). Nickel cadmium battery has long cycle life, overcharge capability, high rates of discharge and charge, almost constant discharge voltage and possibility of operation at low temperature. But, the cost of cadmium is several times that of lead and the cost of nickel cadmium cell construction is more expensive than that of lead acid cell. And there is also problem with the manipulation of toxic cadmium. But also low maintenance and good reliability have made it an ideal for a number of applications such (emergency lighting, engine starting, portable television receivers, hedge trimmers,

Depending on construction, nickel cadmium cells have energy densities in the range 40- 60 Wh/kg (50-150 Wh/dm3). Cycle life is moving from several hundreds for sealed cells to

Cell construction is branched to two types. First using pocket plate electrodes (in vented cells). The active material is found in pockets of finely perforated nickel plated sheet steel. Positive and negative plates are then separated by plastic pins or ladders and plate edge insulators. Second using sintered, bonded or fibre plate electrodes (in both vented and sealed cells). In sintered plate electrodes, a porous sintered nickel electrode is sintered in belt furnace in reducing atmosphere at 800 to 1000°C. Active material is distributed within the pores. In sintered plate cells, a special woven or felted nylon separator is used. It permits oxygen diffusion (oxygen cycle) in sealed cells. In the most common version, a spiral or

The electrolyte is an aqueous solution KOH (concentration of 20-28% by weight and a density of 1.18-1.27 g/cm3 at 25°C). 1-2% of LiOH is usually added to electrolyte to minimize coagulation of the NiOOH electrode during charge/discharge cycling. For low temperature applications, the more concentrated KOH solution is used. When it is operating at high

2NiOOH + Cd + 2H2O → 2Ni(OH)2 + Cd(OH)2 E0 = +1.30 V (2)

It is notable that amount of water in the electrolyte falls during discharge. Ni-Cd batteries are designed as positive limited utilizing oxygen cycle. The oxygen evolved at the positive electrode during charge difuses to the negative electrode and reacts with cadmium to form

## *3.1.1. Battery composition and construction*

The sealed nickel metal hydride cell has with hydrogen absorbed in a metal alloy as the active negative material. When compare with Ni-Cd cell it is not only increases the energy density, but also it is a more environmentally friendly power source. The nickel metal hydride cell, however, has high selfdischarge and is less tolerant to overcharge than the Ni-Cd cell.

Positive electrode is NiOOH, negative electrode contains hydrogen absorption alloys. They can absorb over a thousand times their own volume of hydrogen: Alloys usually consist of two metals. First absorbs hydrogen exothermically, a second endothermically. They serve as a catalyst for the dissociative adsorption of atomic hydrogen into the alloy lattice. Examples of used metals: Pd, V, Ti, Zr, Ni, Cr, Co, Sn, Fe, lanthanides and others. The AB2 series (ZrNi2) and the AB5 series (LaNi5) are usually used.

Design of the cylindrical and prismatic sealed Ni-MH cells are similar as with a nickel cadmium cells (see Fig. 7). Hydrophilic polypropylene separator is used in Ni-MH cell.

**Figure 7.** Scheme of prismatic and spiral wound Ni-MH battery

## *3.1.2. Principle of operation*

The overall reaction during discharge:

$$\text{NiOOH} + \text{MH} \rightarrow \text{Ni(OH)}\_2 + \text{M} \tag{3}$$

Electrochemical Energy Storage 89

After shallow cycling there is a voltage step during discharge, i.e. as if the cell remembers the depth of the shallow cycling. The size of the voltage reduction depends on the number of preceding shallow cycles and the value of the discharge current. But the capacity of the cell is not affected if the cell is now fully discharged (to 0.9 V) and then recharged. Deep discharge then shows a normal discharge curve. It seems that some morphological change occurs in the undischarged active material during the shallow cycling. It could cause a reduction of the cell voltage during folowing discharge. The effect is probably based on an increase in the resistance of the undischarged material (-NiOOH formation on overcharge during the shallow cycles) (Vincent &

Progressive irreversible capacity loss can be confused with the reversible memory effect. The former is caused different mechanisms. For example by a reduction in the electrolyte volume due to evaporation at high temperatures or prolonged overcharge. Irreversible

Lithium is attractive as a battery negative electrode material because it is light weight, high reduction potential and low resistance. Development of high energy density lithium-ion battery started in the 1970s. The lithium-ion cell contains no metallic lithium and is therefore

The principle of the lithium-ion cell is illustrated schematically in Fig. 8. The lithium ions

The most of commercial lithium-ion cells have positive electrodes of cobalt oxide. Other possible positive electrodes are except LiCoO2 and LiNiO2 based on especially manganese

Negative electrode is carbon, in the form of either graphite or an amorphous material with a high surface-area. Carbon is an available and cheap material of low weight and also it is able to absorb a good quantity of lithium. When paired with a metal oxide as the positive electrode it gives a cell with a relatively high voltage (from 4 V in the fully charged state to

Electrolyte is composed from organic liquid (ether) and dissolved salt (LiPF6, LiBF4, LiClO4). The positive and negative active mass is applied to both sides of thin metal foils (aluminium on positive and copper on negative). Microporous polymer sheet between the positive and

Lithium-ion cells are produced in coin format, as well as in cylindrical and prismatic (see

much safer on recharge than the earlier, primary lithium-metal design of cell.

travel between one electrode and the other during charge and discharge.

capacity loss can also be caused by internal short circuits.

*3.2.1. Battery composition and construction* 

oxide, namely, LiMnO2 and LiMn2O4.

3 V in discharged state) (Dell & Rand, 2001).

negative electrode works as the separator.

Fig. 9) shapes.

Scrosati, 2003).

**3.2. Li-ion battery** 

The electrolyte is concentrated potassium hydroxide, voltage is in the range 1.32-1.35 V, depending on used alloy. Water is not involved in the cell reaction.

The energy density is 25% higher than a Ni-Cd cell (80 Wh/kg), power density around 200 W/kg, cycle life over 1000 cycles. Self-discharge is high - up to 4-5% per day. It is caused especially by the hydrogen dissolved in the electrolyte that reacts with the positive electrode.

Ni-MH batteries are used in hybrid electric vehicle batteries, electric razors, toothbrushes, cameras, camcorders, mobile phones, pagers, medical instruments, and numerous other high rate long cycle life applications.

## *3.1.3. Memory effect*

Ni-MH batteries also suffer from the memory effect. It is a reversible process which results in the temporary reduction of the capacity of a Ni-Cd and Ni-MH cell. It is caused by shallow charge-discharge cycling.

After shallow cycling there is a voltage step during discharge, i.e. as if the cell remembers the depth of the shallow cycling. The size of the voltage reduction depends on the number of preceding shallow cycles and the value of the discharge current. But the capacity of the cell is not affected if the cell is now fully discharged (to 0.9 V) and then recharged. Deep discharge then shows a normal discharge curve. It seems that some morphological change occurs in the undischarged active material during the shallow cycling. It could cause a reduction of the cell voltage during folowing discharge. The effect is probably based on an increase in the resistance of the undischarged material (-NiOOH formation on overcharge during the shallow cycles) (Vincent & Scrosati, 2003).

Progressive irreversible capacity loss can be confused with the reversible memory effect. The former is caused different mechanisms. For example by a reduction in the electrolyte volume due to evaporation at high temperatures or prolonged overcharge. Irreversible capacity loss can also be caused by internal short circuits.

## **3.2. Li-ion battery**

88 Energy Storage – Technologies and Applications

**Figure 7.** Scheme of prismatic and spiral wound Ni-MH battery

depending on used alloy. Water is not involved in the cell reaction.

NiOOH + MH → Ni(OH)2 + M (3)

The electrolyte is concentrated potassium hydroxide, voltage is in the range 1.32-1.35 V,

The energy density is 25% higher than a Ni-Cd cell (80 Wh/kg), power density around 200 W/kg, cycle life over 1000 cycles. Self-discharge is high - up to 4-5% per day. It is caused especially by the hydrogen dissolved in the electrolyte that reacts with the positive electrode.

Ni-MH batteries are used in hybrid electric vehicle batteries, electric razors, toothbrushes, cameras, camcorders, mobile phones, pagers, medical instruments, and numerous other

Ni-MH batteries also suffer from the memory effect. It is a reversible process which results in the temporary reduction of the capacity of a Ni-Cd and Ni-MH cell. It is caused by

*3.1.2. Principle of operation* 

The overall reaction during discharge:

high rate long cycle life applications.

shallow charge-discharge cycling.

*3.1.3. Memory effect* 

Lithium is attractive as a battery negative electrode material because it is light weight, high reduction potential and low resistance. Development of high energy density lithium-ion battery started in the 1970s. The lithium-ion cell contains no metallic lithium and is therefore much safer on recharge than the earlier, primary lithium-metal design of cell.

## *3.2.1. Battery composition and construction*

The principle of the lithium-ion cell is illustrated schematically in Fig. 8. The lithium ions travel between one electrode and the other during charge and discharge.

The most of commercial lithium-ion cells have positive electrodes of cobalt oxide. Other possible positive electrodes are except LiCoO2 and LiNiO2 based on especially manganese oxide, namely, LiMnO2 and LiMn2O4.

Negative electrode is carbon, in the form of either graphite or an amorphous material with a high surface-area. Carbon is an available and cheap material of low weight and also it is able to absorb a good quantity of lithium. When paired with a metal oxide as the positive electrode it gives a cell with a relatively high voltage (from 4 V in the fully charged state to 3 V in discharged state) (Dell & Rand, 2001).

Electrolyte is composed from organic liquid (ether) and dissolved salt (LiPF6, LiBF4, LiClO4). The positive and negative active mass is applied to both sides of thin metal foils (aluminium on positive and copper on negative). Microporous polymer sheet between the positive and negative electrode works as the separator.

Lithium-ion cells are produced in coin format, as well as in cylindrical and prismatic (see Fig. 9) shapes.

Electrochemical Energy Storage 91

The most important advantages of lithium-ion cell are high energy density from 150 to 200 Wh/kg (from 250 to 530 Wh/l), high voltage (3.6 V), good charge-discharge characteristics, with more than 500 cycles possible, acceptably low selfdischarge (< 10% per month), absence of a memory effect, much safer than equivalent cells which use lithium

Main disadvantage is a high price of the lithium-ion battery. There also must be controlled charging process, especially close the top of charge voltage 4.2 V. Overcharging or heating above 100°C cause the decomposition of the positive electrode with liberation of oxygen gas

Polymers contained a hetero-atom (i.e. oxygen or sulfur) is able to dissolve lithium salts in very high concentrations. Some experiments were made with polyethylene oxide (PEO), which dissolves salts lithium perchlorate LiClO4 and lithium trifluoromethane sulfonate LiCF3SO3 very well. But there is disadvantage - the conductivity of the solid solution of lithium ions is too low (about 10-5 S/m) on room temperature. But when higher temperature is reached (more than 60°C), transformation of crystalline to amorphous phase proceeds. It leads to much better electrical conductivity (10-1 S/m at 100°C). This value allows the polymer to serve as an electrolyte for lithium batteries. But thickness of the polymer must be low (10 to 100 m). Polymer electrolyte is safer then liquid electrolyte, because it is not

The zinc-silver oxide battery has one of the highest energy of aqueous cells. The theoretical energy density is 300 Wh/kg (1400 Wh/dm3) and practical values are in the range 40- 130 Wh/kg (110-320 Wh/dm3). Cells have poor cycle life. But they can reach a very low internal resistance and also their high energy density makes them very useful for aerospace

The silver positive active mass is formed by sintering of silver powder at temperatures between 400 and 700°C and it is placed on silver or silver-plated copper grids or perforated

The zinc negative electrode prepares as mixtures of zinc, zinc oxide and organic binding agents. The aim is to produce electrodes of high porosity. Other additives include surface active agents to minimize dendritic growth and mercuric ions to increase the hydrogen overvoltage of the zinc electrode (reduce gassing during charge) and so reduce corrosion.

metal, possibility of rapid recharging (2h).

(LiCoO2 yields Co3O4).

*3.2.3. Li-pol battery* 

flammable (Dell & Rand, 2001).

*4.1.1. Battery composition and construction* 

Electrolyte is water solution of KOH (1.40 to 1.42 g/cm3).

**4. Special batteries** 

and even military purposes.

sheets.

**4.1. Ag-Zn battery** 

**Figure 8.** The principle of the lithium-ion cell

**Figure 9.** Prismatic and cylindrical Li-ion cell construction

#### *3.2.2. Principle of operation*

The positive electrode reaction is:

Positive electrode:

$$\text{LiCoO} \rightleftharpoons \text{Li} \cdot \text{CoO} + \text{xLi}^+ + \text{xe} \tag{4}$$

Negative electrode:

$$\text{xLi}^{+} + \text{xe}^{\cdot} + \text{Cu} \begin{array}{c} \neg \text{LiCl} \end{array} \tag{5}$$

where x moves on negative electrode from 0 to 1, on positive electrode from 0 to 0.45.

The most important advantages of lithium-ion cell are high energy density from 150 to 200 Wh/kg (from 250 to 530 Wh/l), high voltage (3.6 V), good charge-discharge characteristics, with more than 500 cycles possible, acceptably low selfdischarge (< 10% per month), absence of a memory effect, much safer than equivalent cells which use lithium metal, possibility of rapid recharging (2h).

Main disadvantage is a high price of the lithium-ion battery. There also must be controlled charging process, especially close the top of charge voltage 4.2 V. Overcharging or heating above 100°C cause the decomposition of the positive electrode with liberation of oxygen gas (LiCoO2 yields Co3O4).

## *3.2.3. Li-pol battery*

90 Energy Storage – Technologies and Applications

**Figure 8.** The principle of the lithium-ion cell

**Figure 9.** Prismatic and cylindrical Li-ion cell construction

LiCoO2 → Li1-xCoO2 + xLi+ + xe-

where x moves on negative electrode from 0 to 1, on positive electrode from 0 to 0.45.

(4)

+ C6 → LixC6 , (5)

*3.2.2. Principle of operation* 

Positive electrode:

Negative electrode:

The positive electrode reaction is:

xLi+ + xe-

Polymers contained a hetero-atom (i.e. oxygen or sulfur) is able to dissolve lithium salts in very high concentrations. Some experiments were made with polyethylene oxide (PEO), which dissolves salts lithium perchlorate LiClO4 and lithium trifluoromethane sulfonate LiCF3SO3 very well. But there is disadvantage - the conductivity of the solid solution of lithium ions is too low (about 10-5 S/m) on room temperature. But when higher temperature is reached (more than 60°C), transformation of crystalline to amorphous phase proceeds. It leads to much better electrical conductivity (10-1 S/m at 100°C). This value allows the polymer to serve as an electrolyte for lithium batteries. But thickness of the polymer must be low (10 to 100 m). Polymer electrolyte is safer then liquid electrolyte, because it is not flammable (Dell & Rand, 2001).

## **4. Special batteries**

## **4.1. Ag-Zn battery**

## *4.1.1. Battery composition and construction*

The zinc-silver oxide battery has one of the highest energy of aqueous cells. The theoretical energy density is 300 Wh/kg (1400 Wh/dm3) and practical values are in the range 40- 130 Wh/kg (110-320 Wh/dm3). Cells have poor cycle life. But they can reach a very low internal resistance and also their high energy density makes them very useful for aerospace and even military purposes.

The silver positive active mass is formed by sintering of silver powder at temperatures between 400 and 700°C and it is placed on silver or silver-plated copper grids or perforated sheets.

The zinc negative electrode prepares as mixtures of zinc, zinc oxide and organic binding agents. The aim is to produce electrodes of high porosity. Other additives include surface active agents to minimize dendritic growth and mercuric ions to increase the hydrogen overvoltage of the zinc electrode (reduce gassing during charge) and so reduce corrosion.

Electrolyte is water solution of KOH (1.40 to 1.42 g/cm3).

The separator is the most important component of zinc-silver oxide cell. It must prevent short circuit between electrodes, must prevent silver migration to the negative electrode, to control zincate migration, to preserve the integrity of the zinc electrode. The separator must have a low ion resistance with good thermal and chemical stability in KOH solution. Typical separators used in Ag-Zn battery, are of cellophane (regenerated cellulose), synthetic fiber mats of nylon, polypropylene, and nonwoven rayon fiber mats. Synthetic fiber mats are placed next to the positive electrode to protect the cellophane from oxidizing influence of that material. In most cells the separators are in form of envelopes completely enclosed the zinc electrodes (Vincent & Scrosati, 2003).

Electrochemical Energy Storage 93

During discharge there rises metal silver inside positive electrode and that is why inner electrical resistance drops in discharged state. Maximum temperature range is from -40 to

Zinc-silver oxide secondary cells with capacities of 0.5-100 Ah are manufactured for use in space satellites, military aircraft, submarines and for supplying power to portable military equipment. In space applications the batteries are used to increase the power from solar cells during period of high demand, e.g. during radio transmission or when the sun is eclipsed.

The Ni-H2 battery is an alkaline battery developed especially for use in satellites (see

Fig. 11). It is a hybrid battery combining battery and fuel cell technology.

50 °C. Self discharge of Ag-Zn battery at 25 °C is about 4% of capacity per month.

At other times the batteries are charged by the solar cells.

**Figure 11.** Scheme of a nickel-hydrogen battery (Zimmerman, 2009)

*4.2.1. Battery composition and construction* 

**4.2. Ni-H2 battery** 

Commercial cells are generally prismatic – see Fig. 10 in shape and the case is usually plastic. Construction must be able to withstand the mechanical stress. The cells are usually sealed with safety valves. The volume of free electrolyte is very small. It is absorbed in the electrode pores and separator.

The energy density of practical zinc-silver oxide cells is some five to six times higher than that of their nickel-cadmium cells. The main disadvantage of the system is its high cost combined with a poor cycle life.

**Figure 10.** Ag-Zn prismatic and submarine torpedo battery

### *4.1.2. Principle of operation*

The overall cell reaction during discharge:

$$\text{AgBr} + 2\text{HxO} + 2\text{Zn} \quad \begin{array}{c} \neg 2\text{Ag} + 2\text{Zn(OH)} \text{n} \end{array} \tag{6}$$

The cell discharge reaction takes place in two stages:

$$\text{Ag} \cdot \text{O} + \text{HO} + 2\text{Zn} \longrightarrow \text{Ag} \cdot \text{O} + 2\text{Zn(OH)}\text{n} \qquad \qquad \text{E}^{\text{0}} = +1.85 \text{ V} \tag{7}$$

$$\text{AgSO} + \text{HO} + \text{Zn} \quad \begin{array}{c} \rightharpoonup \text{2Ag} + \text{Zn(OH)} \text{:} \end{array} \text{H} \begin{array}{c} \text{0} = \text{+1.59 V} \end{array} \tag{8}$$

During discharge there rises metal silver inside positive electrode and that is why inner electrical resistance drops in discharged state. Maximum temperature range is from -40 to 50 °C. Self discharge of Ag-Zn battery at 25 °C is about 4% of capacity per month.

Zinc-silver oxide secondary cells with capacities of 0.5-100 Ah are manufactured for use in space satellites, military aircraft, submarines and for supplying power to portable military equipment. In space applications the batteries are used to increase the power from solar cells during period of high demand, e.g. during radio transmission or when the sun is eclipsed. At other times the batteries are charged by the solar cells.

## **4.2. Ni-H2 battery**

92 Energy Storage – Technologies and Applications

zinc electrodes (Vincent & Scrosati, 2003).

electrode pores and separator.

combined with a poor cycle life.

*4.1.2. Principle of operation* 

**Figure 10.** Ag-Zn prismatic and submarine torpedo battery

The cell discharge reaction takes place in two stages:

The overall cell reaction during discharge:

The separator is the most important component of zinc-silver oxide cell. It must prevent short circuit between electrodes, must prevent silver migration to the negative electrode, to control zincate migration, to preserve the integrity of the zinc electrode. The separator must have a low ion resistance with good thermal and chemical stability in KOH solution. Typical separators used in Ag-Zn battery, are of cellophane (regenerated cellulose), synthetic fiber mats of nylon, polypropylene, and nonwoven rayon fiber mats. Synthetic fiber mats are placed next to the positive electrode to protect the cellophane from oxidizing influence of that material. In most cells the separators are in form of envelopes completely enclosed the

Commercial cells are generally prismatic – see Fig. 10 in shape and the case is usually plastic. Construction must be able to withstand the mechanical stress. The cells are usually sealed with safety valves. The volume of free electrolyte is very small. It is absorbed in the

The energy density of practical zinc-silver oxide cells is some five to six times higher than that of their nickel-cadmium cells. The main disadvantage of the system is its high cost

Ag2O2 + 2H2O + 2Zn → 2Ag + 2Zn(OH)2 (6)

Ag2O2 + H2O + 2Zn → Ag2O + 2Zn(OH)2 E0 = +1.85 V (7)

Ag2O + H2O + Zn → 2Ag + Zn(OH)2 E0 = +1.59 V (8)

### *4.2.1. Battery composition and construction*

The Ni-H2 battery is an alkaline battery developed especially for use in satellites (see Fig. 11). It is a hybrid battery combining battery and fuel cell technology.

**Figure 11.** Scheme of a nickel-hydrogen battery (Zimmerman, 2009)

The battery has a sintered, nickel-oxide positive electrode and a negative electrode from platinum black catalyst supported with Teflon bonding dispersed on carbon paper. Two types of the separators are being used. First is formed from a porous ceramic paper, made from fibres of yttria-stabilized zirconia, second from asbestos paper (Linden & Reddy, 2002). Separators absorb the potassium hydroxide electrolyte. The battery was developed to replace Ni-Cd battery in space applications and it has some higher specific energy (50 Wh/kg) together with a very long cycle life. Standard voltage of the Ni-H2 cell is 1.32 V.

## *4.2.2. Principle of operation*

The overall reaction during discharge:

$$\text{2NiOOH} + \text{H}\_2 \xrightarrow{} \text{2Ni(OH)\_2} \tag{9}$$

Electrochemical Energy Storage 95

**Figure 12.** Scheme of zinc-bromine battery (Dell & Rand, 2001)

Zn + Br2 → ZnBr2 E0 = +1.85 V (10)

During discharge product of reaction, the soluble zinc bromide is stored, along with the rest of the electrolyte, in the two loops and external tanks. During charge, bromine is liberated on the positive electrode and zinc is deposited on the negative electrode. Bromine is then complexed with an organic agent to form a dense, oily liquid polybromide complex. It is produced as droplets and these are separated from the aqueous electrolyte on the bottom of the tank in positive electrode loop. During discharge, bromine in positive electrode loop is again returned to the cell electrolyte in the form of a dispersion of the polybromide oil.

A vanadium redox battery is another type of a flow battery in which electrolytes in two loops are separated by a proton exchange membrane (PEM). The electrolyte is prepared by dissolving of vanadium pentoxide (V2O5) in sulphuric acid (H2SO4). The electrolyte in the positive electrolyte loop contains (VO2)+ - (V5+) and (VO)2+ - (V4+) ions, the electrolyte in the negative electrolyte loop, V3+ and V2+ ions. Chemical reactions proceed on the carbon

The overall chemical reaction during discharge:

*5.1.2. Principle of operation* 

**5.2. Vanadium redox battery** 

electrodes.

*5.2.1. Battery composition and construction* 

The hydrogen gas liberated on charging is stored under pressure within the cell pressure vessel. Shape of the vessel is cylindrical with hemi-spherical end caps made from thin, Inconel alloy. Pressure of hydrogen inside the vessel grows to 4 MPa during charge whereas in the discharged state falls to 0.2 MPa. The cells may be overcharged because liberated oxygen from the positive electrode recombines rapidly at the negative electrode into the water.

## **5. Flow batteries**

Flow batteries store and release electrical energy with help of reversible electrochemical reactions in two liquid electrolytes. An electrochemical cell has two loops physically separated by an ion or proton exchange membrane. Electrolytes flow into and out of the cell through separate loops and undergo chemical reaction inside the cell, with ion or proton exchange through the membrane and electron exchange through the external electric circuit. There are some advantages to using the flow battery when compared with a conventional secondary battery. The capacity of the system is possible to scale by increasing the amount of solution in electrolyte tanks. The battery can be fully discharged and has little loss of electrolyte during cycling. Because the electrolytes are stored separately, flow batteries have a low selfdischarge. Disadvantage is a low energy density and specific energy.

## **5.1. Br2-Zn battery**

## *5.1.1. Battery composition and construction*

The zinc-bromine cell is composed from the bipolar electrodes. The bipolar electrode is from a lightweight, carbon-plastic composite material. Microporous plastic separator between electrodes allows the ions to pass through it. Cells are series-connected and the battery has a positive and a negative electrode loop. The electrolyte in each storage tank is circulated through the appropriate loop.

**Figure 12.** Scheme of zinc-bromine battery (Dell & Rand, 2001)

#### *5.1.2. Principle of operation*

94 Energy Storage – Technologies and Applications

*4.2.2. Principle of operation* 

The overall reaction during discharge:

1.32 V.

water.

**5. Flow batteries** 

**5.1. Br2-Zn battery** 

through the appropriate loop.

*5.1.1. Battery composition and construction* 

The battery has a sintered, nickel-oxide positive electrode and a negative electrode from platinum black catalyst supported with Teflon bonding dispersed on carbon paper. Two types of the separators are being used. First is formed from a porous ceramic paper, made from fibres of yttria-stabilized zirconia, second from asbestos paper (Linden & Reddy, 2002). Separators absorb the potassium hydroxide electrolyte. The battery was developed to replace Ni-Cd battery in space applications and it has some higher specific energy (50 Wh/kg) together with a very long cycle life. Standard voltage of the Ni-H2 cell is

The hydrogen gas liberated on charging is stored under pressure within the cell pressure vessel. Shape of the vessel is cylindrical with hemi-spherical end caps made from thin, Inconel alloy. Pressure of hydrogen inside the vessel grows to 4 MPa during charge whereas in the discharged state falls to 0.2 MPa. The cells may be overcharged because liberated oxygen from the positive electrode recombines rapidly at the negative electrode into the

Flow batteries store and release electrical energy with help of reversible electrochemical reactions in two liquid electrolytes. An electrochemical cell has two loops physically separated by an ion or proton exchange membrane. Electrolytes flow into and out of the cell through separate loops and undergo chemical reaction inside the cell, with ion or proton exchange through the membrane and electron exchange through the external electric circuit. There are some advantages to using the flow battery when compared with a conventional secondary battery. The capacity of the system is possible to scale by increasing the amount of solution in electrolyte tanks. The battery can be fully discharged and has little loss of electrolyte during cycling. Because the electrolytes are stored separately, flow batteries have

The zinc-bromine cell is composed from the bipolar electrodes. The bipolar electrode is from a lightweight, carbon-plastic composite material. Microporous plastic separator between electrodes allows the ions to pass through it. Cells are series-connected and the battery has a positive and a negative electrode loop. The electrolyte in each storage tank is circulated

a low selfdischarge. Disadvantage is a low energy density and specific energy.

2NiOOH + H2 → 2Ni(OH)2 (9)

The overall chemical reaction during discharge:

$$\text{Zn} + \text{Br}\_2 \xrightarrow{\text{\tiny}} \text{ZnBr}\_2 \qquad\qquad \text{E}^\text{\text{\tiny}} = +1.85 \text{ V} \tag{10}$$

During discharge product of reaction, the soluble zinc bromide is stored, along with the rest of the electrolyte, in the two loops and external tanks. During charge, bromine is liberated on the positive electrode and zinc is deposited on the negative electrode. Bromine is then complexed with an organic agent to form a dense, oily liquid polybromide complex. It is produced as droplets and these are separated from the aqueous electrolyte on the bottom of the tank in positive electrode loop. During discharge, bromine in positive electrode loop is again returned to the cell electrolyte in the form of a dispersion of the polybromide oil.

#### **5.2. Vanadium redox battery**

#### *5.2.1. Battery composition and construction*

A vanadium redox battery is another type of a flow battery in which electrolytes in two loops are separated by a proton exchange membrane (PEM). The electrolyte is prepared by dissolving of vanadium pentoxide (V2O5) in sulphuric acid (H2SO4). The electrolyte in the positive electrolyte loop contains (VO2)+ - (V5+) and (VO)2+ - (V4+) ions, the electrolyte in the negative electrolyte loop, V3+ and V2+ ions. Chemical reactions proceed on the carbon electrodes.

**Figure 13.** Scheme of vanadium redox battery

## *5.2.2. Principle of operation*

In the vanadium redox cell, the following half-cell reactions are involved during discharge:

At the negative electrode:

$$\mathbf{V}^{\sharp \star} \rightharpoonup \mathbf{V}^{\sharp \star} + \mathbf{e}^{\flat} \tag{11}$$

Electrochemical Energy Storage 97

material which can be used in combination with sodium to form a cell. Sulphur is also

The problem of a sodium-sulphur cell is to find a suitable electrolyte. Aqueous electrolytes cannot be used and, unlike the lithium, no suitable polymer was found. That is why a ceramic material beta-alumina (–Al2O3) was used as electrolyte. It is an electronic insulator,

In each cell, the negative electrode (molten sodium) was contained in a vertical tube (diameter from 1 to 2 cm). The positive electrode (molten sulphur) is absorbed into the pores of carbon felt (serves as the current-collector) and inserted into the annulus between the ceramic beta-alumina electrolyte tube and the cylindrical steel case (Fig. 14). Between molten sodium and beta-alumina electrolyte also could be found a safety liner with a pin-hole in its

The cell discharges at 300 to 400 °C. Sodium ions pass from the sodium negative electrode, through the beta-alumina electrolyte, to the sulphur positive electrode. There they react with

3Na2S5 + 4Na → 5Na2S3 E0 = 1.78V (14)

Uncontrolled chemical reaction of molten sodium and sulphur could cause a fire and corrosion inside the cell and consequently destruction of the cell. It often happens after the fracture of the electrolyte tube. This problem is solved by inserting of safety liner to the beta-

5S + 2Na → Na2S5 E0 = 2.076V (13)

the sulphur to form sodium polysulphides. Standard voltage of the cell is about 2 V.

highly available in nature and very cheap.

base.

but above 300 °C it has a high ionic conductivity for sodium ions.

**Figure 14.** Schematic cross-section of Na-S cell (Dell & Rand, 2001)

*6.1.2. Principle of operation* 

The cell discharges in two steps:

At the positive electrode:

$$\rm VO^{+} + 2H^{+} + e^{-} \rightarrow VO^{2+} + H\_{2}O \qquad \qquad \quad \rightarrow^{0} = 1.00 \text{V} \tag{12}$$

Under actual cell conditions, an open circuit voltage of 1.4 Volts is observed at 50% state of charge, while a fully charged cell produces over 1.6 Volts at open-circuit, fully discharged cell 1.0 Volt.

The extremely large capacities possible from vanadium redox batteries make them well suited to use in large RAPS applications, where they could to average out the production of highly unstable power sources such as wind or solar power. The extremely rapid response times make them suitable for UPS type applications, where they can be used to replace lead acid batteries. Disadvantage of vanadium redox batteries is a low energy density of about 25 Wh/kg of electrolyte, low charge efficiency (necessity using of pumps) and a high price.

## **6. High temperature batteries**

### **6.1. Na-S battery**

### *6.1.1. Battery composition and construction*

Sodium, just like lithium, has many advantages as a negative-electrode material. Sodium has a high reduction potential of -2.71V and a low atomic weight (23.0). These properties allow to made a battery with a high specific energy (100-200 Wh/kg). Sodium salts are highly found in nature, they are cheap and non-toxic. Sulphur is the positive electrode material which can be used in combination with sodium to form a cell. Sulphur is also highly available in nature and very cheap.

The problem of a sodium-sulphur cell is to find a suitable electrolyte. Aqueous electrolytes cannot be used and, unlike the lithium, no suitable polymer was found. That is why a ceramic material beta-alumina (–Al2O3) was used as electrolyte. It is an electronic insulator, but above 300 °C it has a high ionic conductivity for sodium ions.

In each cell, the negative electrode (molten sodium) was contained in a vertical tube (diameter from 1 to 2 cm). The positive electrode (molten sulphur) is absorbed into the pores of carbon felt (serves as the current-collector) and inserted into the annulus between the ceramic beta-alumina electrolyte tube and the cylindrical steel case (Fig. 14). Between molten sodium and beta-alumina electrolyte also could be found a safety liner with a pin-hole in its base.

**Figure 14.** Schematic cross-section of Na-S cell (Dell & Rand, 2001)

#### *6.1.2. Principle of operation*

96 Energy Storage – Technologies and Applications

**Figure 13.** Scheme of vanadium redox battery

In the vanadium redox cell, the following half-cell reactions are involved during discharge:

VO2+ + 2H+ + e- → VO2+ + H2O E0 = 1.00V (12)

Under actual cell conditions, an open circuit voltage of 1.4 Volts is observed at 50% state of charge, while a fully charged cell produces over 1.6 Volts at open-circuit, fully discharged

The extremely large capacities possible from vanadium redox batteries make them well suited to use in large RAPS applications, where they could to average out the production of highly unstable power sources such as wind or solar power. The extremely rapid response times make them suitable for UPS type applications, where they can be used to replace lead acid batteries. Disadvantage of vanadium redox batteries is a low energy density of about 25 Wh/kg of electrolyte, low charge efficiency (necessity using of pumps) and a high price.

Sodium, just like lithium, has many advantages as a negative-electrode material. Sodium has a high reduction potential of -2.71V and a low atomic weight (23.0). These properties allow to made a battery with a high specific energy (100-200 Wh/kg). Sodium salts are highly found in nature, they are cheap and non-toxic. Sulphur is the positive electrode

E0 = -0.26V (11)

*5.2.2. Principle of operation* 

At the negative electrode:

At the positive electrode:

cell 1.0 Volt.

**6.1. Na-S battery** 

V2+ → V3+ + e-

**6. High temperature batteries** 

*6.1.1. Battery composition and construction* 

The cell discharges at 300 to 400 °C. Sodium ions pass from the sodium negative electrode, through the beta-alumina electrolyte, to the sulphur positive electrode. There they react with the sulphur to form sodium polysulphides. Standard voltage of the cell is about 2 V.

The cell discharges in two steps:

$$\text{\tiny 5S + 2Na \rightarrow Na\text{\tiny 5S}}\tag{13}$$

3Na2S5 + 4Na → 5Na2S3 E0 = 1.78V (14)

Uncontrolled chemical reaction of molten sodium and sulphur could cause a fire and corrosion inside the cell and consequently destruction of the cell. It often happens after the fracture of the electrolyte tube. This problem is solved by inserting of safety liner to the beta-

alumina tube. This allows a normal flow of sodium to the inner wall of the beta-alumina electrolyte, but prevents the flow in the case of tube fracture.

Electrochemical Energy Storage 99

Advantage of the sodium metalchloride cell over the sodium sulphur cell is that there is possibility of both an overcharge and overdischarge reaction, when the second electrolyte (molten sodium chloraluminate) reacts with metal (overcharge) or with sodium

2NaAlCl4 + Ni → 2Na + 2AlCl3 + NiCl2 (17)

Another advantage of the sodium metalchloride system is safety of operation. When the beta-alumina electrolyte tube cracks in this system, the molten sodium first encounters the

This chapter is focused on electrochemical storage or batteries that constitute a large group of technologies that are potentially suitable to meet a broad market needs. The five categories of electrochemical systems (secondary batteries) were selected and discussed in detail: standard batteries (lead acid, Ni-Cd) modern batteries (Ni-MH, Li–ion, Li-pol), special batteries (Ag-Zn, Ni-H2), flow batteries (Br2-Zn, vanadium redox) and high temperature batteries (Na-S, Na–metalchloride). These batteries appear to be promising to

However, the use of secondary batteries involves some technical problems. Since their cells slowly self-discharge, batteries are mostly suitable for electricity storage only for limited

For electrochemical energy storage, the specific energy and specific power are two important parameters. Other important parameters are ability to charge and discharge a large number of times, to retain charge as long time as possible and ability to charge and

This chapter is supported by the EU project CZ.1.05/2.1.00/01.0014 and by the internal grant

periods of time. They also age, which results in a decreasing storage capacity.

NaAlCl4 electrolyte and reacts with it according the overdischarge reaction.

3Na +NaAlCl4 → Al + 4NaCl (18)

(overdischarge).

**7. Conclusion** 

**Author details** 

Petr Krivik and Petr Baca

**Acknowledgement** 

FEKT-S-11-7.

Overcharge reaction for sodium nickelchloride cell:

Overdischarge reaction for sodium nickelchloride cell:

meet the requirements for end-user applications.

discharge over a wide range of temperatures.

*Brno University of Technology, Czech Republic* 

*The Faculty of Electrical Engineering and Communication,* 

## **6.2. Na-metalchloride battery**

## *6.2.1. Battery composition and construction*

In the sodium-metalchloride battery the sulphur positive electrode there is replaced by nickel chloride or by a mixture of nickel chloride (NiCl2) and ferrous chloride (FeCl2) – see Fig. 15. The specific energy is 100-200 Wh/kg.

The negative electrode is from molten sodium, positive electrode from metalchloride and electrolyte from the ceramic beta-alumina (the same as in the sodium-sulphur battery). The second electrolyte, to make good ionic contact between the positive electrode and the electrolyte from beta-alumina, is molten sodium chloraluminate (NaAlCl4).

The positive electrode is from a mixture of metal powder (Ni or Fe) and sodium chloride (NaCl). During charge, these materials are converted into the corresponding metal chloride and sodium. Iron powder is cheaper than nickel powder, but nickel cells have higher voltage and could operate over a wider temperature range (200 to 400 °C) than iron cells (200 to 300 °C) (Dell & Rand, 2001).

**Figure 15.** Schematic cross-section of Na-metalchloride cell (Rand, 1998)

## *6.2.2. Principle of operation*

The basic cell reactions during discharge are simple, i.e.

$$\begin{array}{cc} \text{2Na} + \text{NiCl} \rightarrow \text{2NaCl} + \text{Ni} & \begin{array}{c} \text{E}^{\circ} = +2.58 \text{ V} \end{array} \end{array} \tag{15}$$

$$\text{2Na} + \text{FeCl} \rightleftharpoons \text{2NaCl} + \text{Fe} \tag{16}$$

Advantage of the sodium metalchloride cell over the sodium sulphur cell is that there is possibility of both an overcharge and overdischarge reaction, when the second electrolyte (molten sodium chloraluminate) reacts with metal (overcharge) or with sodium (overdischarge).

Overcharge reaction for sodium nickelchloride cell:

$$2\text{NaAlCl} + \text{Ni} \xrightarrow{} 2\text{Na} + 2\text{AlCl} \text{J} + \text{NiCl} \tag{17}$$

Overdischarge reaction for sodium nickelchloride cell:

$$\text{AlNa} + \text{NaAlCl} \rightleftharpoons \text{Al} + \text{NaCl} \tag{18}$$

Another advantage of the sodium metalchloride system is safety of operation. When the beta-alumina electrolyte tube cracks in this system, the molten sodium first encounters the NaAlCl4 electrolyte and reacts with it according the overdischarge reaction.

## **7. Conclusion**

98 Energy Storage – Technologies and Applications

**6.2. Na-metalchloride battery** 

(200 to 300 °C) (Dell & Rand, 2001).

*6.2.2. Principle of operation* 

*6.2.1. Battery composition and construction* 

Fig. 15. The specific energy is 100-200 Wh/kg.

electrolyte, but prevents the flow in the case of tube fracture.

alumina tube. This allows a normal flow of sodium to the inner wall of the beta-alumina

In the sodium-metalchloride battery the sulphur positive electrode there is replaced by nickel chloride or by a mixture of nickel chloride (NiCl2) and ferrous chloride (FeCl2) – see

The negative electrode is from molten sodium, positive electrode from metalchloride and electrolyte from the ceramic beta-alumina (the same as in the sodium-sulphur battery). The second electrolyte, to make good ionic contact between the positive electrode and the

The positive electrode is from a mixture of metal powder (Ni or Fe) and sodium chloride (NaCl). During charge, these materials are converted into the corresponding metal chloride and sodium. Iron powder is cheaper than nickel powder, but nickel cells have higher voltage and could operate over a wider temperature range (200 to 400 °C) than iron cells

2Na + NiCl2 → 2NaCl + Ni E0 = +2.58 V (15)

2Na + FeCl2 → 2NaCl + Fe E0 = +2.35 V (16)

electrolyte from beta-alumina, is molten sodium chloraluminate (NaAlCl4).

**Figure 15.** Schematic cross-section of Na-metalchloride cell (Rand, 1998)

The basic cell reactions during discharge are simple, i.e.

This chapter is focused on electrochemical storage or batteries that constitute a large group of technologies that are potentially suitable to meet a broad market needs. The five categories of electrochemical systems (secondary batteries) were selected and discussed in detail: standard batteries (lead acid, Ni-Cd) modern batteries (Ni-MH, Li–ion, Li-pol), special batteries (Ag-Zn, Ni-H2), flow batteries (Br2-Zn, vanadium redox) and high temperature batteries (Na-S, Na–metalchloride). These batteries appear to be promising to meet the requirements for end-user applications.

However, the use of secondary batteries involves some technical problems. Since their cells slowly self-discharge, batteries are mostly suitable for electricity storage only for limited periods of time. They also age, which results in a decreasing storage capacity.

For electrochemical energy storage, the specific energy and specific power are two important parameters. Other important parameters are ability to charge and discharge a large number of times, to retain charge as long time as possible and ability to charge and discharge over a wide range of temperatures.

## **Author details**

Petr Krivik and Petr Baca *The Faculty of Electrical Engineering and Communication, Brno University of Technology, Czech Republic* 

## **Acknowledgement**

This chapter is supported by the EU project CZ.1.05/2.1.00/01.0014 and by the internal grant FEKT-S-11-7.

### **8. References**

Calabek, M. et al. (2001). A fundamental study of the effects of compression on the performance of lead accumulator plates, *J. Power Sources*, Vol. 95, 97 – 107, ISSN 0378- 7753

**Chapter 4** 

© 2013 Chen et al., licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

© 2013 Chen et al., licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

**Compressed Air Energy Storage** 

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/52221

**1. Introduction** 

Haisheng Chen, Xinjing Zhang, Jinchao Liu and Chunqing Tan

European countries, and even higher for Japan in the near future[4][10].

Electrical Energy Storage (EES) refers to a process of converting electrical energy from a power network into a form that can be stored for converting back to electrical energy when needed [1-3]. Such a process enables electricity to be produced at times of either low demand, low generation cost or from intermittent energy sources and to be used at times of high demand, high generation cost or when no other generation is available[1-9].The history of EES dates back to the turn of 20th century, when power stations often shut down for overnight, with lead-acid accumulators supplying the residual loads on the then direct current (DC) networks [2-4]. Utility companies eventually recognised the importance of the flexibility that energy storage provides in networks and the first central station energy storage, a Pumped Hydroelectric Storage (PHS), was in use in 1929[2][10-15]. Up to 2011, a total of more than 128 GW of EES has been installed all over the world [9-12]. EES systems is currently enjoying somewhat of a renaissance, for a variety of reasons including changes in the worldwide utility regulatory environment, an ever-increasing reliance on electricity in industry, commerce and the home, power quality/quality-of-supply issues, the growth of renewable energy as a major new source of electricity supply, and all combined with ever more stringent environmental requirements[3- 4][6]. These factors, combined with the rapidly accelerating rate of technological development in many of the emerging electrical energy storage systems, with anticipated unit cost reductions, now make their practical applications look very attractive on future timescales of only years. The anticipated storage level will boost to 10~15% of delivered inventory for USA and

There are numerous EES technologies including Pumped Hydroelectric Storage (PHS)[11- 12][17], Compressed Air Energy Storage system (CAES)[18-22], Battery[23-27], Flow Battery[3-4][6][13], Fuel Cell[24][28], Solar Fuel[4][29], Superconducting Magnetic Energy Storage system (SMES)[30- 32], Flywheel[13][16][33-34] and Capacitor and Supercapacitor[4][16]. However, only two kinds of EES technologies are credible for energy storage in large scale (above 100MW in single unit) i.e. PHS and CAES. PHS is the most widely implemented large-scale form of EES. Its


## **Compressed Air Energy Storage**

Haisheng Chen, Xinjing Zhang, Jinchao Liu and Chunqing Tan

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/52221

## **1. Introduction**

100 Energy Storage – Technologies and Applications

ISBN 0-85404-605-4, Cambridge, UK

Fe Drive, Denver, Colorado, USA

Plaza, ISBN 0-07-135978-8, New York, USA

86380-205-2, Taunton, Somerset, Great Britain

*Sources,* Vol. 127, 33–44, ISSN 0378-7753

Calabek, M. et al. (2001). A fundamental study of the effects of compression on the performance of lead accumulator plates, *J. Power Sources*, Vol. 95, 97 – 107, ISSN 0378-

Dell, R.M. & Rand, D.A.J. (2001). *Understanding Batteries*, The Royal Society of Chemistry,

Linden, D. & Reddy, T.B. (2002). *Handbook of Batteries, Third Edition*, McGraw-Hill, Two Penn

Nelson, R. (2001). The Basic Chemistry of Gas Recombination in Lead-Acid Batteries, Santa

Rand, D.A.J. et al. (1998). *Batteries for Electric Vehicles*, Research Studies Press Ltd., ISBN 0-

Rand, D.A.J. et al. (2004). *Valve-regulated Lead-Acid Batteries*, Elsevier B.V., ISBN 0-444-50746-

Ruetschi, P. (2004). Aging mechanisms and service life of lead–acid batteries. *J. Power* 

Vincent, C.A. & Scrosati, B. (2003). *Modern Batteries*, Antony Rowe Ltd, ISBN 0-340-66278-6,

Zimmerman, A.(2009). Nickel-Hydrogen Batteries: Principles and Practice, Available from

http://www.aero.org/publications/zimmerman/chapter2.html

**8. References** 

7753

9, Netherlands

Eastbourne, Great Britain

Electrical Energy Storage (EES) refers to a process of converting electrical energy from a power network into a form that can be stored for converting back to electrical energy when needed [1-3]. Such a process enables electricity to be produced at times of either low demand, low generation cost or from intermittent energy sources and to be used at times of high demand, high generation cost or when no other generation is available[1-9].The history of EES dates back to the turn of 20th century, when power stations often shut down for overnight, with lead-acid accumulators supplying the residual loads on the then direct current (DC) networks [2-4]. Utility companies eventually recognised the importance of the flexibility that energy storage provides in networks and the first central station energy storage, a Pumped Hydroelectric Storage (PHS), was in use in 1929[2][10-15]. Up to 2011, a total of more than 128 GW of EES has been installed all over the world [9-12]. EES systems is currently enjoying somewhat of a renaissance, for a variety of reasons including changes in the worldwide utility regulatory environment, an ever-increasing reliance on electricity in industry, commerce and the home, power quality/quality-of-supply issues, the growth of renewable energy as a major new source of electricity supply, and all combined with ever more stringent environmental requirements[3- 4][6]. These factors, combined with the rapidly accelerating rate of technological development in many of the emerging electrical energy storage systems, with anticipated unit cost reductions, now make their practical applications look very attractive on future timescales of only years. The anticipated storage level will boost to 10~15% of delivered inventory for USA and European countries, and even higher for Japan in the near future[4][10].

There are numerous EES technologies including Pumped Hydroelectric Storage (PHS)[11- 12][17], Compressed Air Energy Storage system (CAES)[18-22], Battery[23-27], Flow Battery[3-4][6][13], Fuel Cell[24][28], Solar Fuel[4][29], Superconducting Magnetic Energy Storage system (SMES)[30- 32], Flywheel[13][16][33-34] and Capacitor and Supercapacitor[4][16]. However, only two kinds of EES technologies are credible for energy storage in large scale (above 100MW in single unit) i.e. PHS and CAES. PHS is the most widely implemented large-scale form of EES. Its

© 2013 Chen et al., licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2013 Chen et al., licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

principle is to store hydraulic potential energy by pumping water from a lower reservoir to an elevated reservoir. PHS is a mature technology with large volume, long storage period, high efficiency and relatively low capital cost per unit energy. However, it has a major drawback of the scarcity of available sites for two large reservoirs and one or two dams. A long lead time (typically ~10 years) and a large amount of cost (typically hundreds to thousands million US dollars) for construction and environmental issues (e.g. removing trees and vegetation from the large amounts of land prior to the reservoir being flooded) are the other three major constrains in the deployment of PHS. These drawbacks or constrains of PHS make CAES an attracting alternative for large scale energy storage. CAES is the only other commercially available technology (besides the PHS) able to provide the very-large system energy storage deliverability (above 100MW in single unit) to use for commodity storage or other large-scale storage.

Compressed Air Energy Storage 103

3. The turbine train, containing both high- and low pressure turbines.

**Figure 1.** Schematic diagram of gas turbine and CAES system

when adding or releasing air.

facility.

compressed air.

4. Equipment controls for operating the combustion turbine, compressor, and auxiliaries and to regulate and control changeover from generation mode to storage mode. 5. Auxiliary equipment consisting of fuel storage and handling, and mechanical and electrical systems for various heat exchangers required to support the operation of the

6. The under-ground component is mainly the cavity used for the storage of the

The storage cavity can potentially be developed in three different categories of geologic formations: underground rock caverns created by excavating comparatively hard and impervious rock formations; salt caverns created by solution- or dry-mining of salt formations; and porous media reservoirs made by water-bearing aquifers or depleted gas or oil fields (for example, sandstone, fissured lime). Aquifers in particular can be very attractive as storage media because the compressed air will displace water, setting up a constant pressure storage system while the pressure in the alternative systems will vary

The chapter aims to review research and application state-of-arts of CAES including principle, function and deployments. The chapter is structured in the following manner. Section 2 will give the principle of CAES. Technical characteristics of the CAES will be described in Section 3 in terms of power rating and discharge time, storage duration, energy efficiency, energy density, cycle life and life time, capital cost etc. Functions and deployments will be given in Sections 4 and 5. And research and development of new CAES technologies will be discussed in Section 6. Finally, concluding remarks will be made in Section 7.

## **2. Principle**

The concept of CAES can be dated back to 1949 when Stal Laval filed the first patent of CAES which used an underground cavern to store the compressed air[9]. Its principle is on the basis of conventional gas turbine generation. As shown in Figure 1, CAES decouples the compression and expansion cycle of a conventional gas turbine into two separated processes and stores the energy in the form of the elastic potential energy of compressed air. In low demand period, energy is stored by compressing air in an air tight space (typically 4.0~8.0 MPa) such as underground storage cavern. To extract the stored energy, compressed air is drawn from the storage vessel, mixed with fuel and combusted, and then expanded through a turbine. And the turbine is connected to a generator to produce electricity. The waste heat of the exhaust can be captured through a recuperator before being released to the atmosphere (figure 2).

As shown in Figure 2, a CAES system is made of above-ground and below-ground components that combine man-made technology and natural geological formations to accept, store, and dispatch energy. There are six major components in a basic CAES installation including five above-ground and one under-ground components:


3. The turbine train, containing both high- and low pressure turbines.

102 Energy Storage – Technologies and Applications

storage or other large-scale storage.

**2. Principle** 

atmosphere (figure 2).

air.

compressor or turbine trains.

principle is to store hydraulic potential energy by pumping water from a lower reservoir to an elevated reservoir. PHS is a mature technology with large volume, long storage period, high efficiency and relatively low capital cost per unit energy. However, it has a major drawback of the scarcity of available sites for two large reservoirs and one or two dams. A long lead time (typically ~10 years) and a large amount of cost (typically hundreds to thousands million US dollars) for construction and environmental issues (e.g. removing trees and vegetation from the large amounts of land prior to the reservoir being flooded) are the other three major constrains in the deployment of PHS. These drawbacks or constrains of PHS make CAES an attracting alternative for large scale energy storage. CAES is the only other commercially available technology (besides the PHS) able to provide the very-large system energy storage deliverability (above 100MW in single unit) to use for commodity

The chapter aims to review research and application state-of-arts of CAES including principle, function and deployments. The chapter is structured in the following manner. Section 2 will give the principle of CAES. Technical characteristics of the CAES will be described in Section 3 in terms of power rating and discharge time, storage duration, energy efficiency, energy density, cycle life and life time, capital cost etc. Functions and deployments will be given in Sections 4 and 5. And research and development of new CAES technologies

The concept of CAES can be dated back to 1949 when Stal Laval filed the first patent of CAES which used an underground cavern to store the compressed air[9]. Its principle is on the basis of conventional gas turbine generation. As shown in Figure 1, CAES decouples the compression and expansion cycle of a conventional gas turbine into two separated processes and stores the energy in the form of the elastic potential energy of compressed air. In low demand period, energy is stored by compressing air in an air tight space (typically 4.0~8.0 MPa) such as underground storage cavern. To extract the stored energy, compressed air is drawn from the storage vessel, mixed with fuel and combusted, and then expanded through a turbine. And the turbine is connected to a generator to produce electricity. The waste heat of the exhaust can be captured through a recuperator before being released to the

As shown in Figure 2, a CAES system is made of above-ground and below-ground components that combine man-made technology and natural geological formations to accept, store, and dispatch energy. There are six major components in a basic CAES

1. The motor/generator that employs clutches to provide for alternate engagement to the

2. The air compressor that may require two or more stages, intercoolers and after-coolers, to achieve economy of compression and reduce the moisture content of the compressed

installation including five above-ground and one under-ground components:

will be discussed in Section 6. Finally, concluding remarks will be made in Section 7.


**Figure 1.** Schematic diagram of gas turbine and CAES system

The storage cavity can potentially be developed in three different categories of geologic formations: underground rock caverns created by excavating comparatively hard and impervious rock formations; salt caverns created by solution- or dry-mining of salt formations; and porous media reservoirs made by water-bearing aquifers or depleted gas or oil fields (for example, sandstone, fissured lime). Aquifers in particular can be very attractive as storage media because the compressed air will displace water, setting up a constant pressure storage system while the pressure in the alternative systems will vary when adding or releasing air.

Compressed Air Energy Storage 105

**Figure 3.** Technical characteristics of CAES

work at full power. As a result, CAES has following functions

CAES systems are designed to cycle on a daily basis and to operate efficiently during partial load conditions. This design approach allows CAES units to swing quickly from generation to compression modes. CAES plant can start without extra power input and take minutes to

**4. Function** 

**Figure 2.** Components of CAES[35]

## **3. Technical characteristics**

Figure 3 shows the comparison of technical characteristics between CAES and other EES technologies. One can see that CAES has a long storage period, low capital costs but relatively low efficiency. The typical ratings for a CAES system are in the range 50 to 300 MW and currently manufacturers can create CAES machinery for facilities ranging from 5 to 350 MW. The rating is much higher than for other storage technologies other than pumped hydro. The storage period is also longer than other storage methods since the losses are very small; actually a CAES system can be used to store energy for more than a year. The typical value of storage efficiency of CAES is in the range of 60-80%. Capital costs for CAES facilities vary depending on the type of underground storage but are typically in the range from \$400 to \$800 per kW. The typical specific energy density is 3-6 Wh/litre or 0.5-2 W/litre and the typical life time is 20-40 years.

Similar to PHS, the major barrier to implementation of CAES is also the reliance on favourable geography such as caverns hence is only economically feasible for power plants that have nearby rock mines, salt caverns, aquifers or depleted gas fields. In addition, in comparison with PHS and other currently available energy storage systems, CAES is not an independent system and requires to be associated the gas turbine plant. It cannot be used in other types of power plants such as coal-fired, nuclear, wind turbine or solar photovoltaic plants. More importantly, the combustion of fossil fuel leads to emission of contaminates such as nitrogen oxides and carbon oxide which render the CAES less attractive[19][36,37]. Many improved CAES are proposed or under investigation, for example Small Scale CAES with fabricated small vessels and **A**dvanced **A**diabatic CAES (AACAES) with TES[19][21], which will be discussed in Section 6.

**Figure 3.** Technical characteristics of CAES

#### **4. Function**

104 Energy Storage – Technologies and Applications

**Figure 2.** Components of CAES[35]

**3. Technical characteristics** 

and the typical life time is 20-40 years.

which will be discussed in Section 6.

Figure 3 shows the comparison of technical characteristics between CAES and other EES technologies. One can see that CAES has a long storage period, low capital costs but relatively low efficiency. The typical ratings for a CAES system are in the range 50 to 300 MW and currently manufacturers can create CAES machinery for facilities ranging from 5 to 350 MW. The rating is much higher than for other storage technologies other than pumped hydro. The storage period is also longer than other storage methods since the losses are very small; actually a CAES system can be used to store energy for more than a year. The typical value of storage efficiency of CAES is in the range of 60-80%. Capital costs for CAES facilities vary depending on the type of underground storage but are typically in the range from \$400 to \$800 per kW. The typical specific energy density is 3-6 Wh/litre or 0.5-2 W/litre

Similar to PHS, the major barrier to implementation of CAES is also the reliance on favourable geography such as caverns hence is only economically feasible for power plants that have nearby rock mines, salt caverns, aquifers or depleted gas fields. In addition, in comparison with PHS and other currently available energy storage systems, CAES is not an independent system and requires to be associated the gas turbine plant. It cannot be used in other types of power plants such as coal-fired, nuclear, wind turbine or solar photovoltaic plants. More importantly, the combustion of fossil fuel leads to emission of contaminates such as nitrogen oxides and carbon oxide which render the CAES less attractive[19][36,37]. Many improved CAES are proposed or under investigation, for example Small Scale CAES with fabricated small vessels and **A**dvanced **A**diabatic CAES (AACAES) with TES[19][21],

CAES systems are designed to cycle on a daily basis and to operate efficiently during partial load conditions. This design approach allows CAES units to swing quickly from generation to compression modes. CAES plant can start without extra power input and take minutes to work at full power. As a result, CAES has following functions

1. Peak shaving: Utility systems that benefit from the CAES include those with load varying significantly during the daily cycle and with costs varying significantly with the generation level or time of day. It is economically important that storing and moving low-cost power into higher price markets, reducing peak power prices.

Compressed Air Energy Storage 107

2. Project Markham, Texas: This 540 MW project developed jointly by Ridege Energy Services and EI Paso Energy will consist of four 135 MW CAES units with separate low pressure and high pressure motor driven compression trains. A salt dome is used as the

3. Iowa stored energy project: This project under development by Iowa Association of Municipal Utilities, promises to be exciting and innovative. The compressed air will be stored in an underground aquifer, and wind energy will be used to compress air, in addition to available off-peak power. The plant configuration is for 200MW of CAES generating capacity, with 100MW of wind energy. CAES will expand the role of wind energy in the region generation mix, and will operate to follow loads and provide capacity when other generation is unavailable or non-economic. The underground aquifer near Fort Dodge has the ideal dome structure allowing large volumes of air

4. Japan Chubu project: Chubu Electric of Japan is surveying its service territory for appropriate CAES sites. Chubu is Japan's third largest electric utility with 14 thermal and two nuclear power plants that generate 21,380 MWh of electicity annually. Japanese utilities recognize the value of storing off-peak power in a nation where peak electricity

5. Eskom project: Eskom of South Africa has expressed interest in exploring the economic

As mentioned in Section 3, there are two major barriers to implementation of CAES: the reliance on favourable caverns and the reliance on fossil fuel. To alleviate the barriers, many improved CAES systems are proposed or under research and development, typical examples are improved conventional CAES, **A**dvanced **A**diabatic CAES (AACAES) with

Figure 4 shows the principle of the improved conventional CAES system, which is similar to Figure 3. In figure 4, there are intercoolers and aftercooler in the compression process; reheater is installed between turbine stages; and regenerator is used to preheat the compressed air by the exhausted gas. McIntosh plant can reduce fuel consumption by 25%

Another improved conventional CAES system combined with a gas turbine is shown in figure 5[38-41]. When the electricity is in low-demand, the compressed air is produced and stored in underground cavity or above ground reservoir. During the high-demand period, the CAES is charging the grid simultaneously with the GT power system. The compressed air is heated by the GT exhaustion and the heated compression air expands in the high pressure (HP) turbine and then ejects to the GT turbine combustor to join GT working fluid.

The CAES system shown in figure 4 can recover almost 70% of compression energy.

benefits of CAES in one of its integrated energy plans[10].

TES[19][21] and Small Scale CAES with fabricated small vessels.

**6.1. Improved conventional CAES system** 

using the improved cycle shown in figure 4.

storage vessel.

storage at 3.6 MPa pressure.

costs can reach \$0.53/kWh.

**6. Research and development** 


## **5. Deployment**

Although CAES is a mature, commercially available energy storage technology, there are only two CAES operated all over the world. One is in Huntorf in Germany, another is in Mclntosh, Alabama in USA. The CAES plant in Huntorf, Germany is the oldest operating CAES system. It has been in operation for about 30 years since 1978. The Huntorf CAES system is a 290 MW, 50Hz unit, owned and operated by the Nordwestdeutche Krafiwerke, AG. The size of the cavern, which is located in a solution mined salt dome about 600m underground, is approximately 310,000 m3. It runs on a daily cycle with eight hours of charging required to fill the cavern. Operating flexibility, however, is greatly limited by the small cavern size. Compression is achieved through the use of electrically driven 60 MW compressors up to a maximum pressure of 10 MPa. At full load the plant can generate 290 MW for two hours. Since its installation, the plant has showed high operation ability e.g. 90% availability and 99% starting reliability.

The second commercial CAES plant, owned by the Alabama Energy Cooperative (AEC) in McIntosh, Alabama, has been in operation for more than 15 years since 1991. The CAES system stores compressed air with a pressure of up to 7.5 MPa in an underground cavern located in a solution mined salt dome 450m below the surface. The storage capacity is over 500,000 m3 with a generating capacity of 110 MW. Natural gas heats the air released from the cavern, which is then expanded through a turbine to generate electricity. It can provide 26 hours of generation. The McIntosh CAES system utilizes a recuperator to reuse heat energy from the gas turbine, which reduces fuel consumption by 25% compared with the Huntorf CAES plant.

There are several planed or under development CAES projects:

1. The third commercial CAES is a 2700 MW plant that is planned for construction in the United States at Norton, Ohio developed by Haddington Ventures Inc.. This 9-unit plant will compress air to ~10 MPa in an existing limestone mine dome 670m under ground. The volume of the storage cavern is about 120,000,000 m3.


## **6. Research and development**

106 Energy Storage – Technologies and Applications

(demand) charges.

90% availability and 99% starting reliability.

There are several planed or under development CAES projects:

ground. The volume of the storage cavern is about 120,000,000 m3.

**5. Deployment** 

Huntorf CAES plant.

1. Peak shaving: Utility systems that benefit from the CAES include those with load varying significantly during the daily cycle and with costs varying significantly with the generation level or time of day. It is economically important that storing and

3. Energy Management: CAES allows customers to peak shave by shifting energy demand from one time of the day to another. This is primarily used to reduce their time-of-use

4. Renewable energy: Linking CAES systems to intermittent renewable resources, it can

5. Standby power: CAES could also replace conventional battery system as a standby

Although CAES is a mature, commercially available energy storage technology, there are only two CAES operated all over the world. One is in Huntorf in Germany, another is in Mclntosh, Alabama in USA. The CAES plant in Huntorf, Germany is the oldest operating CAES system. It has been in operation for about 30 years since 1978. The Huntorf CAES system is a 290 MW, 50Hz unit, owned and operated by the Nordwestdeutche Krafiwerke, AG. The size of the cavern, which is located in a solution mined salt dome about 600m underground, is approximately 310,000 m3. It runs on a daily cycle with eight hours of charging required to fill the cavern. Operating flexibility, however, is greatly limited by the small cavern size. Compression is achieved through the use of electrically driven 60 MW compressors up to a maximum pressure of 10 MPa. At full load the plant can generate 290 MW for two hours. Since its installation, the plant has showed high operation ability e.g.

The second commercial CAES plant, owned by the Alabama Energy Cooperative (AEC) in McIntosh, Alabama, has been in operation for more than 15 years since 1991. The CAES system stores compressed air with a pressure of up to 7.5 MPa in an underground cavern located in a solution mined salt dome 450m below the surface. The storage capacity is over 500,000 m3 with a generating capacity of 110 MW. Natural gas heats the air released from the cavern, which is then expanded through a turbine to generate electricity. It can provide 26 hours of generation. The McIntosh CAES system utilizes a recuperator to reuse heat energy from the gas turbine, which reduces fuel consumption by 25% compared with the

1. The third commercial CAES is a 2700 MW plant that is planned for construction in the United States at Norton, Ohio developed by Haddington Ventures Inc.. This 9-unit plant will compress air to ~10 MPa in an existing limestone mine dome 670m under

moving low-cost power into higher price markets, reducing peak power prices. 2. Load leveling: CAES plants can respond to load changes to provide load following

because they are designed to sustain frequent start-up/shut-down cycles.

increase the capacity credit and improve environmental characteristics.

power which decreases the construction and operation time and cost.

As mentioned in Section 3, there are two major barriers to implementation of CAES: the reliance on favourable caverns and the reliance on fossil fuel. To alleviate the barriers, many improved CAES systems are proposed or under research and development, typical examples are improved conventional CAES, **A**dvanced **A**diabatic CAES (AACAES) with TES[19][21] and Small Scale CAES with fabricated small vessels.

## **6.1. Improved conventional CAES system**

Figure 4 shows the principle of the improved conventional CAES system, which is similar to Figure 3. In figure 4, there are intercoolers and aftercooler in the compression process; reheater is installed between turbine stages; and regenerator is used to preheat the compressed air by the exhausted gas. McIntosh plant can reduce fuel consumption by 25% using the improved cycle shown in figure 4.

Another improved conventional CAES system combined with a gas turbine is shown in figure 5[38-41]. When the electricity is in low-demand, the compressed air is produced and stored in underground cavity or above ground reservoir. During the high-demand period, the CAES is charging the grid simultaneously with the GT power system. The compressed air is heated by the GT exhaustion and the heated compression air expands in the high pressure (HP) turbine and then ejects to the GT turbine combustor to join GT working fluid. The CAES system shown in figure 4 can recover almost 70% of compression energy.

Compressed Air Energy Storage 109

**Figure 6.** Schematic diagram of AA-CAES system

Small-scale CAES system (<~10MW) with man-made vessels is a more adaptable solution, without need of caverns, especially for distributed generation that could be widely applicable to future power networks. Figure 7 shows a small-scale CAES used for standby power system[42]. It can replace battery with technical simplicity, low degradation of components, high reliability, low maintenance and lower life cycle cost characteristics. For a 2kW power application, CAES can work 20 years, while vented lead acid batteries (VLAB) 12 years; the installation and commissioning durations are 8 hours, respectively, while 16 and 64 hours for VLAB; with 300bar, 24,000L compressed air in cylinders, the CAES can work as a standby power for one year by charging four times. In general, there is no heat recovery/storage component in the small-scale CAES system, therefore its efficiency is lower

**Figure 7.** Schematic diagram of the CAES system as a standby power supply

**6.3. Small-scale CAES System** 

than that of VLAB system.

**Figure 4.** Schematic diagrams of improved conventional CAES system

**Figure 5.** CAES combined with GT system

#### **6.2. Advanced Adiabatic CAES system**

The so called Advanced Adiabatic CAES (AA-CAES) stores the potential and thermal energy of compressed air separately, and recover them during expansion (as shown in figure 6). Although the cost is about 20~30% higher than the conventional power plant, this system eliminates the combustor and is a fossil free system. IAA-CAES may be commercially viable due to the improvements of thermal energy storage (TES), compressor and turbine technologies. A project "AA-CAES" (Advanced Adiabatic – Compressed Air Energy Storage: EC DGXII contract ENK6 CT-2002-00611) committed to developing this technology to meet the current requirements of energy storage.

**Figure 6.** Schematic diagram of AA-CAES system

#### **6.3. Small-scale CAES System**

108 Energy Storage – Technologies and Applications

**Figure 5.** CAES combined with GT system

**6.2. Advanced Adiabatic CAES system** 

technology to meet the current requirements of energy storage.

The so called Advanced Adiabatic CAES (AA-CAES) stores the potential and thermal energy of compressed air separately, and recover them during expansion (as shown in figure 6). Although the cost is about 20~30% higher than the conventional power plant, this system eliminates the combustor and is a fossil free system. IAA-CAES may be commercially viable due to the improvements of thermal energy storage (TES), compressor and turbine technologies. A project "AA-CAES" (Advanced Adiabatic – Compressed Air Energy Storage: EC DGXII contract ENK6 CT-2002-00611) committed to developing this

**Figure 4.** Schematic diagrams of improved conventional CAES system

Small-scale CAES system (<~10MW) with man-made vessels is a more adaptable solution, without need of caverns, especially for distributed generation that could be widely applicable to future power networks. Figure 7 shows a small-scale CAES used for standby power system[42]. It can replace battery with technical simplicity, low degradation of components, high reliability, low maintenance and lower life cycle cost characteristics. For a 2kW power application, CAES can work 20 years, while vented lead acid batteries (VLAB) 12 years; the installation and commissioning durations are 8 hours, respectively, while 16 and 64 hours for VLAB; with 300bar, 24,000L compressed air in cylinders, the CAES can work as a standby power for one year by charging four times. In general, there is no heat recovery/storage component in the small-scale CAES system, therefore its efficiency is lower than that of VLAB system.

## **7. Concluding remarks**

Research and application state-of-arts of compressed air energy storage system are discussed in this chapter including principle, function, deployment and R&D status. CAES is the only other commercially available technology (besides the PHS) able to provide the very-large system energy storage deliverability (above 100MW in single unit). It has a long storage period, low capital costs but relatively low efficiency in comparison with other energy storage technologies. CAES can be used for peak shaving, load leveling, energy management, renewable energy and standby power. However, there are two major barriers to implementation of CAES: the reliance on favourable caverns and the reliance on fossil fuel. To alleviate the barriers, many improved CAES systems are under research and development such as improved conventional CAES, AACAES and Small Scale CAES.

Compressed Air Energy Storage 111

[9] Chen, H., Cong, T. N., Yang, W., Tan, C., Li, Y. and Ding, Y. (2009) Progress in electrical energy storage system: A critical review. Progress in Natural Science, 19:291-312. [10] Ahearne J. (2004) Storage of electric energy, Report on research and development of

[12] Linden S. (2003) "The Commercial World of Energy Storage: A Review of Operating Facilities (under construction or planned)", presentation at the 1st Annual Conference

[13] Linden S. (2006) Bulk energy storage potential in the USA, current developments and

[14] Makansi J. and Abboud J. (2002) Energy storage, the missing link in the electricity value

[15] Akhil A., Swaminathan S, Sen R.K. (1997) Cost analysis of energy storage systems for electric utility applications, Sandia Report, SAND97-0443 UC-1350, Sandia National

[16] Kondoh J., Ishii I., Yamaguchi H., Murata A. (2000) Electrical energy storage systems for

[17] Bueno C. and Carta J.A. (2006), Wind powered pumped hydro storage systems, a means of increasing the penetration of renewable energy in the Canary Islands, Renewable and

[18] Najjar Y. S. H., Zaamout M. S. (1998) Performance analysis of compressed air energy storage (CAES) plant for dry regions, Energy Conversion and Management, vol 39, 1503-1511 [19] Sears J. R. (2004) TEX: The next generation of energy storage technology, Telecommunications Energy Conference, INTELEC 2004. 26th Annual International

[20] Najjar Y.S.H and Jubeh N.M. (2006) Comparison of performance of compressed-air energy-storage plant with compressed-air storage with humidification, Proceeding of

[21] Bullough C., Gatzen C., Jakiel C., Koller M., Nowi A. and Zunft S. (2004) Advanced adiabatic compressed air energy storage for the integration of wind energy. Proceedings

[22] Wang S., Chen G., Fang M. and Wang Q. (2006) A new compressed air energy storage

[23] Cook G.M., Spindler W.C. and Grefe G. (1991) Overview of battery power regulation

[24] Chalk S.G., Miller J.F. (2006), Key challenges and recent progress in batteries, fuel cells and hydrogen storage for clean energy systems, Journal of Power Sources, Vol. 159, 73-80 [25] Kashem M.A., Ledwich G. (2007) Energy requirement for distributed energy resources with battery energy storage for voltage support in three-phase distribution lines,

[26] Kluiters E.C., Schmal D., Ter Veen W. R., Posthumus K. (1999) Testing of a sodium/nickel chloride (ZEBRA) battery for electric propulsion of ships and vehicles,

refrigeration system. Energy Conversion and Management, 47, 3408-3416.

and storage. IEEE Transctions on Energy Conversion, vol.6, 204-211

energy networks, Energy Conversion& Management, vol 41, 1863-1874

IMechE, Part A: Journal of Power and Energy, vol. 220, 581-588

of the Eruopean Wind Energy Conference. London UK.

Electric Power Systems Research, vol. 77, 10-23

Journal of Power Sources, vol 80, 261-264

energy technologies. IUPAP working group on energy, 76-86

of the Energy Storage Council, Houston, Texas, 3 March 2003.

future prospects, Energy, vol.31, 3446-3457

Sustainable Energy Reviews, Vol. 10, 312-340

Volume, Issue, 19-23 Sept. 2004, 218 – 222

Laboratories

chain, An ESC White Paper, Energy storage Council

[11] http://en.wikipeida.org/wiki/Hydroelectric\_energy\_storage, 20 March 2007.

## **Author details**

Haisheng Chen\* , Xinjing Zhang, Jinchao Liu and Chunqing Tan *Institute of Engineering Thermophysics, Chinese Academy of Sciences, Beijing, 100190, China* 

## **Acknowledgement**

The authors thank the Natural Science Foundation of China under grant No. 50906079 and Beijing Natural Science Foundation under grant No.3122033 for financial supports.

## **8. References**


<sup>\*</sup> Corresponding Author


Research and application state-of-arts of compressed air energy storage system are discussed in this chapter including principle, function, deployment and R&D status. CAES is the only other commercially available technology (besides the PHS) able to provide the very-large system energy storage deliverability (above 100MW in single unit). It has a long storage period, low capital costs but relatively low efficiency in comparison with other energy storage technologies. CAES can be used for peak shaving, load leveling, energy management, renewable energy and standby power. However, there are two major barriers to implementation of CAES: the reliance on favourable caverns and the reliance on fossil fuel. To alleviate the barriers, many improved CAES systems are under research and

development such as improved conventional CAES, AACAES and Small Scale CAES.

The authors thank the Natural Science Foundation of China under grant No. 50906079 and

[1] Mclarnon F. R., Cairns E. J. (1989) Energy storage, Annul Review of Energy, vol 14, 241-271 [2] Baker J.N. and Collinson A. (1999) Electrical energy storage at the turn of the

[3] Dti Report (2004) Status of electrical energy storage systems, DG/DTI/00050/00/00, URN

[4] Australian Greenhouse Office (2005) Advanced electricity storage technologies

[5] Walawalkar R., Apt J., Mancini R. (2007) Economics of electric energy storage for energy

[6] Dti Report (2004) Review of electrical energy storage technologies and systems and of

[7] Weinstock I. B. (2002) Recent advances in the US department of Energy's energy storage technology research and development programs for hybrid electric and electric

[8] Koot M., Kessels J.T.B.A., Jager B., Heemels W.P.M.H., Bosch P.P. J. and Steinbuch M. (2005) Energy management strategies for vehiclular electric power systems, IEEE

their potential for the UK, DG/DTI/00055/00/00, URN NUMBER 04/1876

, Xinjing Zhang, Jinchao Liu and Chunqing Tan *Institute of Engineering Thermophysics, Chinese Academy of Sciences, Beijing, 100190, China* 

Beijing Natural Science Foundation under grant No.3122033 for financial supports.

Millennium, Power Engineering Journal, No.6, 107-112

arbitrage and regulation, Energy Policy, vol. 35, 2558-2568

vehicles, Journal of Power Sources, vol. 110, 471-474

Transactions on Vehiclular Technology, vol. 54, 771-782

**7. Concluding remarks** 

**Author details** 

**Acknowledgement** 

NUMBER 04/1878

programme, ISBN:1 921120 37 1

Haisheng Chen\*

**8. References** 

 \*

Corresponding Author

	- [27] Karpinski A.P., Makovetski B., Russell S. J., Serenyi J. R., Williams D. C. (1999) Silverzinc: status of technology and applications, vol 80, 53-60

**Chapter 1**

**Chapter 5**

**The Future of Energy Storage Systems**

During the past several years, we have witnessed a radical evolution of electronic devices. One of the major trends of this evolution has been increased portability. Laptops and smart-phone are the most common examples but also cameras or new technologies such as tablets are equal important. The request of efficient energy storage system becomes even more important if we extend it to different applications such as electrical/hybrid vehicles that require hundreds of times larger power when compared with smaller device. Unfortunately the technological improvements of batteries are slower than electronics, creating a constantly growing gap that need to be filled. For this reasons it is very important to develop an efficient energy storage system that goes beyond normal batteries. In the first part of this chapter we will give a general overview of some existing solution such as electrolytic batteries, fuel cells and microturbines. In the second part we will introduce an evolution of simple capacitors known as Supercapacitors or Ultracapacitors. This technology is very promising and it might be able to substitute, or at least improve in a considerable way current energy storage systems.

Nanotechnologies (NTs) can play an important role to help to overcome to energy-related challenges and opportunities. However, what specific kinds of nanotechnologies and how can they provide such advantages? Sepeur [56] defines nanotechnologies as *"'the systematic manipulation production or alteration of structure systems materials or components in the range of atomic and molecular dimension with/into nanoscale dimensions between 1nm and 100nm"'*. In particular two subfields of NTs are interesting for energy problems: Nanofabrication and nanomaterials. By combining these two techniques we are for example able to create structures with a large surface area per unit mass, and by selective etching and deposition of different material layers we are able to fabricate very complex mechanical structures. Furthermore the use of new materials in the process allow us to create films and layers with a specific characteristic (such as conductivity, stress distribution, mechanical resistance etc). By combining nanomaterials and nanofabrication it has for example been possible to build solar cells much more efficient compared to the standard type, to build new classes of materials such

> ©2013 Petricca et al., licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly

©2013 Petricca et al., licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Luca Petricca, Per Ohlckers and Xuyuan Chen

http://dx.doi.org/10.5772/52413

**1. Introduction**

Additional information is available at the end of the chapter

**1.1. Nanotechnologies for energy related issues**

cited.


## **The Future of Energy Storage Systems**

Luca Petricca, Per Ohlckers and Xuyuan Chen

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/52413

## **1. Introduction**

112 Energy Storage – Technologies and Applications

and Technology, 19, R31-R39.

Management, vol 45, 2153-2172

March 2007

Vegas, NV.

Canada.

30th International, pp. 1-4.

storage system, Energy, vol 30, 2128-2143

[27] Karpinski A.P., Makovetski B., Russell S. J., Serenyi J. R., Williams D. C. (1999) Silver-

[28] Weinmann O. (1999) Hydrogen-the flexible storage for electrical energy, Power

[29] Steinfeld A. and Meier A. (2004) Solar thermochemical process technology, in

[30] Kolkert W. J. andJamet F. (1999) Electric energy gun technology: status of the French-German-Netherlands programme, IEEE Transactions on Magnetics, Vol. 35, 25-30 [31] Koshizuka N., Ishikawa F., Nasu H. (2003) Progress of superconducting bearing technologies for flywheel energy storage systems, Physica C, vol. 386, 444-450 [32] Xue X., Cheng K. and Sutanto D. (2006) A study of the status and future of superconducting magnetic energy storage in power systems. Superconductor Science

[33] Suzuki Y., Koyanagi A., Kobayashi M. (2005) Novel applications of the flywheel energy

[34] http://www.beaconpower.com/products/EnergyStorageSystems/ flywheels.htm 20

[35] Jewitt J. (2005) Impact of CAES on Wind in Tx,OK and NM, Presentation in DOE energy storage systems research annual peer review , San Francisco, USA, Oct. 20, 2005 [36] P. Denholm, G. L. Kulcinski. (2004) Life cycle energy requirements and greehouse gas emissions from large scale energy storage systems. Energy Conversion and

[37] P. Denholm, T. Holloway. (2005) Improved accounting of emissions from utility energy storage system operation, Environmental Science & Technology, vol 39, 9016-9022 [38] Nakhamkin, M., Wolk, R. H., Linden, S. v. d. and Patel, M. New Compressed Air Energy Storage Concept Improves the Profitability of Existing Simple Cycle, Combined

[39] Nakhamkin, M. and Chiruvolu, M. (2007) Available Compressed Air Energy Storage

[40] Nakhamkin, M., Chiruvolu, M., Patel, M. and Byrd, S. (2009) Second Generation of CAES Technology-Performance, Operations, Economics, Renewable Load Management, Green Energy. In: POWER-GEN International, Las Vegas Convention Center, Las

[41] Akita, E., Gomi, S., Cloyd, S., Nakhamkin, M. and Chiruvolu, M. (2007) The Air Injection Power Augmentation Technology Provides Additional Significant Operational Benefits. In: ASME Turbo Expo 2007: Power for Land, Sea and Air ASME, Montreal,

[42] Beukes, J., Jacobs, T., Derby, J., Conlon, R. and Henshaw, I. (2008) Suitability of compressed air energy storage technology for electricity utility standby power applications. In: Telecommunications Energy Conference, 2008. INTELEC 2008. IEEE

Cycle, Wind Energy, and Landfill Gas Power Plants. In: ASME, pp. 103-110.

(CAES) Plant Concepts. In: Power-Gen International, Minnestota.

Engineerign Journal, Special Feature: Electrical energy storage, 164-170

Encyclopedia of Energy, Elsevier Inc., Vol. 5, pp. 623-637, 2004.

zinc: status of technology and applications, vol 80, 53-60

During the past several years, we have witnessed a radical evolution of electronic devices. One of the major trends of this evolution has been increased portability. Laptops and smart-phone are the most common examples but also cameras or new technologies such as tablets are equal important. The request of efficient energy storage system becomes even more important if we extend it to different applications such as electrical/hybrid vehicles that require hundreds of times larger power when compared with smaller device. Unfortunately the technological improvements of batteries are slower than electronics, creating a constantly growing gap that need to be filled. For this reasons it is very important to develop an efficient energy storage system that goes beyond normal batteries. In the first part of this chapter we will give a general overview of some existing solution such as electrolytic batteries, fuel cells and microturbines. In the second part we will introduce an evolution of simple capacitors known as Supercapacitors or Ultracapacitors. This technology is very promising and it might be able to substitute, or at least improve in a considerable way current energy storage systems.

## **1.1. Nanotechnologies for energy related issues**

Nanotechnologies (NTs) can play an important role to help to overcome to energy-related challenges and opportunities. However, what specific kinds of nanotechnologies and how can they provide such advantages? Sepeur [56] defines nanotechnologies as *"'the systematic manipulation production or alteration of structure systems materials or components in the range of atomic and molecular dimension with/into nanoscale dimensions between 1nm and 100nm"'*. In particular two subfields of NTs are interesting for energy problems: Nanofabrication and nanomaterials. By combining these two techniques we are for example able to create structures with a large surface area per unit mass, and by selective etching and deposition of different material layers we are able to fabricate very complex mechanical structures. Furthermore the use of new materials in the process allow us to create films and layers with a specific characteristic (such as conductivity, stress distribution, mechanical resistance etc). By combining nanomaterials and nanofabrication it has for example been possible to build solar cells much more efficient compared to the standard type, to build new classes of materials such

©2013 Petricca et al., licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. ©2013 Petricca et al., licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

**Figure 1.** Example of MEMS Energy harvester device [6]

as carbon nanotubes or graphene that are revolutionizing the electronic world. Researchers have already shown that with NT it is possible to create thin film batteries Kuwata et al. [28], Ogawa et al. [44] printable on top of substrates, or creating a smart fibers that can store energy (so called *E-textiles*) Gu et al. [18], Jost et al. [25]. Furthermore, in the literature there are presented many demonstrators of energy harvesting devices that are able to "'harvest"' energy from many different physic sources (such as mechanical vibrations, temperature gradients, electro-magnetic radiations, etc) and transform it into electrical energy.

**Figure 2.** Energy density of different sources (adapted fromPetricca et al. [50])

improving their lifetime.

**2.2. Fuel cells**

air) [37].

which is focused to improve energy density, life time and cycling stability, without using dangerous materials that can create health hazards. Unfortunately it is difficult to find all these proprieties optimized in one material combination. One way for enhancement of the battery capacity is using nanotechnologies for increasing and structuring the surface area of the electrodes (for example by depositing nanomaterials or by growing nanostructures such as nanotubes) [43]. But despite all these efforts, the technological improvements of batteries are still much, much slower when compared with the evolutionary progress of electronics. For these reasons many researchers are trying to include smart circuitry inside the batteries for optimizing the discharge curve by optimizing the load. They call it intelligent batteries [35] and they exploit some battery-related characteristic such as charge recovery effect, for

The Future of Energy Storage Systems 115

Fuel cells are one of the most developed alternatives to batteries and they are already available on the market, with many vehicles currently working on Fuel cells based engines [14]. They are electrochemical devices able to convert the chemical energy stored in the fuels into electrical energy. They mainly consist of two electrodes and a membrane which form a reaction chamber and have external stored reactants. The working principle is similar to batteries, however in this case the species at the electrodes are continuously replenished and they can be refilled, ensuring a continuous electricity supply over a long period. There are many types of fuel cells available on the market, and they mainly differ from the species used as fuel, but all of them use one element as fuel and a second element as oxidizer (commonly

Theoretically with fuel cells we would be able to generate any power or current by changing the physical dimension of the cell and the flow rate of the fuel. However, the voltage across the single cell electrode is fixed and it is not possible to change it. In general this

## **2. Current state of the art**

## **2.1. Batteries**

Nowadays electric batteries represent the most common energy storage methods for portable devices. They store the energy in a chemical way and they are able to reconvert it into electrical form. They consist of two electrodes (anode and cathode) and one electrolyte which can be either solid or liquid; In the redox reaction that powers the battery, reduction occurs at the cathode, while oxidation occurs at the anode [43]. This energy storage form has changed substantially throughout the years, even though the basic principles have been known since the invention by the Italian physicist Alessandro Volta in year 1800. The first type was consisting in a stack of zinc and copper disk separated by an acid electrolyte. Thanks also to the boost of mobile phones during the last years they have evolved to the Nickel-Cadmium (Ni-Ca) and Nickel Metal Hydrate (Ni-Mh) which dominated the market until the developments of Lithium batteries. This latter class rapidly gained market thanks to the higher specific energy (150-500 Wh/Kg versus 50-150Wh/Kg of NiMh, NiCa, See Fig.2) and are nowadays one of the most common batteries available in the market. They can further be divided into another two subclasses which are Lithium Ion (Li-Ion) batteries and Lithium Polymers (Li-Po) batteries (which basically are an evolution of the Li-Ion). The high volume of the market (around 50 billion dollar market in 2006 [39]) is expected to grow even more in the coming years, forecasted to reach around 85.76 billion dollars by 2016 with a Compounded Annual Growth rate around 7% over the next 5 years [34]. This is generating a very high volume of revenues and part of it is re-invested in battery research

**Figure 2.** Energy density of different sources (adapted fromPetricca et al. [50])

which is focused to improve energy density, life time and cycling stability, without using dangerous materials that can create health hazards. Unfortunately it is difficult to find all these proprieties optimized in one material combination. One way for enhancement of the battery capacity is using nanotechnologies for increasing and structuring the surface area of the electrodes (for example by depositing nanomaterials or by growing nanostructures such as nanotubes) [43]. But despite all these efforts, the technological improvements of batteries are still much, much slower when compared with the evolutionary progress of electronics. For these reasons many researchers are trying to include smart circuitry inside the batteries for optimizing the discharge curve by optimizing the load. They call it intelligent batteries [35] and they exploit some battery-related characteristic such as charge recovery effect, for improving their lifetime.

## **2.2. Fuel cells**

2 Energy Storage

as carbon nanotubes or graphene that are revolutionizing the electronic world. Researchers have already shown that with NT it is possible to create thin film batteries Kuwata et al. [28], Ogawa et al. [44] printable on top of substrates, or creating a smart fibers that can store energy (so called *E-textiles*) Gu et al. [18], Jost et al. [25]. Furthermore, in the literature there are presented many demonstrators of energy harvesting devices that are able to "'harvest"' energy from many different physic sources (such as mechanical vibrations, temperature gradients,

Nowadays electric batteries represent the most common energy storage methods for portable devices. They store the energy in a chemical way and they are able to reconvert it into electrical form. They consist of two electrodes (anode and cathode) and one electrolyte which can be either solid or liquid; In the redox reaction that powers the battery, reduction occurs at the cathode, while oxidation occurs at the anode [43]. This energy storage form has changed substantially throughout the years, even though the basic principles have been known since the invention by the Italian physicist Alessandro Volta in year 1800. The first type was consisting in a stack of zinc and copper disk separated by an acid electrolyte. Thanks also to the boost of mobile phones during the last years they have evolved to the Nickel-Cadmium (Ni-Ca) and Nickel Metal Hydrate (Ni-Mh) which dominated the market until the developments of Lithium batteries. This latter class rapidly gained market thanks to the higher specific energy (150-500 Wh/Kg versus 50-150Wh/Kg of NiMh, NiCa, See Fig.2) and are nowadays one of the most common batteries available in the market. They can further be divided into another two subclasses which are Lithium Ion (Li-Ion) batteries and Lithium Polymers (Li-Po) batteries (which basically are an evolution of the Li-Ion). The high volume of the market (around 50 billion dollar market in 2006 [39]) is expected to grow even more in the coming years, forecasted to reach around 85.76 billion dollars by 2016 with a Compounded Annual Growth rate around 7% over the next 5 years [34]. This is generating a very high volume of revenues and part of it is re-invested in battery research

**Figure 1.** Example of MEMS Energy harvester device [6]

**2. Current state of the art**

**2.1. Batteries**

electro-magnetic radiations, etc) and transform it into electrical energy.

Fuel cells are one of the most developed alternatives to batteries and they are already available on the market, with many vehicles currently working on Fuel cells based engines [14]. They are electrochemical devices able to convert the chemical energy stored in the fuels into electrical energy. They mainly consist of two electrodes and a membrane which form a reaction chamber and have external stored reactants. The working principle is similar to batteries, however in this case the species at the electrodes are continuously replenished and they can be refilled, ensuring a continuous electricity supply over a long period. There are many types of fuel cells available on the market, and they mainly differ from the species used as fuel, but all of them use one element as fuel and a second element as oxidizer (commonly air) [37].

Theoretically with fuel cells we would be able to generate any power or current by changing the physical dimension of the cell and the flow rate of the fuel. However, the voltage across the single cell electrode is fixed and it is not possible to change it. In general this

#### 4 Energy Storage 116 Energy Storage – Technologies and Applications The Future of Energy Storage Systems <sup>5</sup>

voltage is very low (less than 1V for realistic operating condition [37]) and thus multiple cell stacks connected in series are needed to achieve larger potentials. Mixed series and parallel connections between different cells can also be used for increase the voltage and the maximum current supplied. Among all the fuel cell types the most promising are the Proton Exchange Membrane (PEM) and the direct Methanol fuel cells (DMFC) which can be considered a PEM special case.

A graphic representation of a PEM fuel cell is shown in Fig.3; In this case the cell use Hydrogen and Oxygen as species. The membrane present between the two electrodes allows passing only the ions while electrons are forced to "'go"' trough the electric circuit.

The chemical reaction at anode is:

$$H\_2 \to 2H^+ + 2e^- \tag{1}$$

researcher from University of Southern California's Loker Hydrocarbon Institute developed a new type of fuel cells which use a direct oxidation of Methanol instead of Hydrogen. Unlike Hydrogen, Methanol offer also the advantage of being liquid at room temperature making storage and transportation much easier. This new class of fuel cells are called Direct Methanol Fuel Cell (DMFC) and the working principle is similar to the PEM fuel cell, but with more

Methanol has energy volume and mass densities of 4380 *Wh*/*l* and 5600 *Wh*/*kg*, which are about 11 times higher than current Li-ion batteries (≈ 500*Wh*/*l*). *"'This means that the FP/FC unit has superior energy density even with an overall conversion efficiency as low as 7%. In the case where no water recycling is employed to minimize system complexity, water has to be carried with methanol for the reforming. With a stoichiometric steam to carbon ratio of 1:1, this reduces the net energy density to 3000 Wh*/*l (4550 Wh*/*kg)"'* [57]. Unfortunately, unlike Hydrogen PEM, where we can assume that all of the polarization losses are located at the cathode, in DMFC the losses at the anode and cathode are comparable. Furthermore DMFC *"'utilizes cathode Pt sites for the direct reaction between methanol and oxygen, which generates a mixed potential that reduces cell voltage"'* [19]. Despite these problems, many companies are already present on the

It has been more than 150 years since the first Internal Combustion Engine (ICE) was developed and nowadays it is the most common power source for vehicles and large engine-generators. However in the last years thanks to the improvements of microtechnologies, many researchers started to design and develop micro internal combustion engines that may be used in the future as power source for small electrical devices. They

forming a few cubic millimeters system able to generate power in mW to Watts range. However, Micro engines are not only a smaller version of the large size counterparts [58]. Some of the technical issue present at this scale are resumed in Fernandez-Pello [13], Sher et al. [58], Suzuki et al. [62], Walther & Ahn [65]. In particular the main challenge is to obtain a genuine combustion in a limited volume of the combustor [45]. Furthermore, at this scale the relative heat losses increase and may cause quenching of the reaction (fuel inside the combustion chamber that rapidly cools down, prevents it from burning) with consequent degradation of performances [13, 45, 65].Moreover, the engine speed and the gap width between the piston and the cylinder walls are two key parameters that can create issues in standard ICE (cylinder-piston engine)as reported by Sher et al. [58]. They simulated the miniaturization limits of a standard ICE with a rotation speed of 48000 rpm, a gap width of 10 um and a compression rate larger than 18 and they found a minimum size limit

<sup>−</sup> + 6*H*<sup>+</sup> (3)

The Future of Energy Storage Systems 117

<sup>−</sup> → 6*H*2*O* (4)

*CH*3*OH* + *H*2*O* → *CO*<sup>2</sup> + 6*e*

12*H*<sup>+</sup> + 6*O*<sup>2</sup> + 12*e*

market offering DMFC [17, 53] or disposable methanol fuel cartridges [64].

**2.3. Micro engines and micro turbines**

consist of three main parts [45],

1. Combustion Chamber

2. Ignition

3. Moving Parts

complicated reactions at the anode (3) and at the cathode (4) [19]:

On the cathode side, the electrons will recombine with the Ions and they will react with the cathode species (Oxygen in this case) forming water through the following reaction:

$$4H^{+} + O\_{2} + 4e^{-} \rightarrow 2H\_{2}O \tag{2}$$

Since water is the only waste product of the reaction, this type of fuel cell is very environmental friendly. Unfortunately there are many problems associated with PEM that are preventing the mass market diffusion of this technology. Hydrogen does not naturally occur in nature and must be produced in factories or laboratories. Furthermore *"'hydrogen has a very high mass energy density (143000 J/g) (See Fig.2) but a very low volumetric energy density (10790 J/L), which makes it difficult to store."'* [50]

**Figure 3.** Fuel cell representation [unknown author]

Other problems associated with PEM are the impurity present in the hydrogen fuel, *"'such as CO, H2S, NH3, organic sulfur carbon, and carbon hydrogen compounds, and in air, such as NOx, SOx, and small organics"',* which are brought in fuel and air feed streams into the electrodes of a PEMFC stack, causing performance degradation or membrane damages [5]. In particular they demonstrated that even small amounts of these impurity materials can poison the anode, cathode or the membrane of the cell, causing a sharp performance drop. To overcome some of these problems (especially the one associated with the hydrogen production and storage), researcher from University of Southern California's Loker Hydrocarbon Institute developed a new type of fuel cells which use a direct oxidation of Methanol instead of Hydrogen. Unlike Hydrogen, Methanol offer also the advantage of being liquid at room temperature making storage and transportation much easier. This new class of fuel cells are called Direct Methanol Fuel Cell (DMFC) and the working principle is similar to the PEM fuel cell, but with more complicated reactions at the anode (3) and at the cathode (4) [19]:

$$\text{CH}\_3\text{OH} + \text{H}\_2\text{O} \rightarrow \text{CO}\_2 + 6e^- + 6H^+ \tag{3}$$

$$12H^{+} + 6O\_{2} + 12e^{-} \rightarrow 6H\_{2}O \tag{4}$$

Methanol has energy volume and mass densities of 4380 *Wh*/*l* and 5600 *Wh*/*kg*, which are about 11 times higher than current Li-ion batteries (≈ 500*Wh*/*l*). *"'This means that the FP/FC unit has superior energy density even with an overall conversion efficiency as low as 7%. In the case where no water recycling is employed to minimize system complexity, water has to be carried with methanol for the reforming. With a stoichiometric steam to carbon ratio of 1:1, this reduces the net energy density to 3000 Wh*/*l (4550 Wh*/*kg)"'* [57]. Unfortunately, unlike Hydrogen PEM, where we can assume that all of the polarization losses are located at the cathode, in DMFC the losses at the anode and cathode are comparable. Furthermore DMFC *"'utilizes cathode Pt sites for the direct reaction between methanol and oxygen, which generates a mixed potential that reduces cell voltage"'* [19]. Despite these problems, many companies are already present on the market offering DMFC [17, 53] or disposable methanol fuel cartridges [64].

## **2.3. Micro engines and micro turbines**

It has been more than 150 years since the first Internal Combustion Engine (ICE) was developed and nowadays it is the most common power source for vehicles and large engine-generators. However in the last years thanks to the improvements of microtechnologies, many researchers started to design and develop micro internal combustion engines that may be used in the future as power source for small electrical devices. They consist of three main parts [45],


4 Energy Storage

voltage is very low (less than 1V for realistic operating condition [37]) and thus multiple cell stacks connected in series are needed to achieve larger potentials. Mixed series and parallel connections between different cells can also be used for increase the voltage and the maximum current supplied. Among all the fuel cell types the most promising are the Proton Exchange Membrane (PEM) and the direct Methanol fuel cells (DMFC) which can be considered a PEM

A graphic representation of a PEM fuel cell is shown in Fig.3; In this case the cell use Hydrogen and Oxygen as species. The membrane present between the two electrodes allows passing

*<sup>H</sup>*<sup>2</sup> <sup>→</sup> <sup>2</sup>*H*<sup>+</sup> <sup>+</sup> <sup>2</sup>*<sup>e</sup>*

On the cathode side, the electrons will recombine with the Ions and they will react with the

Since water is the only waste product of the reaction, this type of fuel cell is very environmental friendly. Unfortunately there are many problems associated with PEM that are preventing the mass market diffusion of this technology. Hydrogen does not naturally occur in nature and must be produced in factories or laboratories. Furthermore *"'hydrogen has a very high mass energy density (143000 J/g) (See Fig.2) but a very low volumetric energy density*

Other problems associated with PEM are the impurity present in the hydrogen fuel, *"'such as CO, H2S, NH3, organic sulfur carbon, and carbon hydrogen compounds, and in air, such as NOx, SOx, and small organics"',* which are brought in fuel and air feed streams into the electrodes of a PEMFC stack, causing performance degradation or membrane damages [5]. In particular they demonstrated that even small amounts of these impurity materials can poison the anode, cathode or the membrane of the cell, causing a sharp performance drop. To overcome some of these problems (especially the one associated with the hydrogen production and storage),

cathode species (Oxygen in this case) forming water through the following reaction:

4*H*<sup>+</sup> + *O*<sup>2</sup> + 4*e*

− (1)

<sup>−</sup> → 2*H*2*O* (2)

only the ions while electrons are forced to "'go"' trough the electric circuit.

special case.

The chemical reaction at anode is:

*(10790 J/L), which makes it difficult to store."'* [50]

**Figure 3.** Fuel cell representation [unknown author]

3. Moving Parts

forming a few cubic millimeters system able to generate power in mW to Watts range. However, Micro engines are not only a smaller version of the large size counterparts [58]. Some of the technical issue present at this scale are resumed in Fernandez-Pello [13], Sher et al. [58], Suzuki et al. [62], Walther & Ahn [65]. In particular the main challenge is to obtain a genuine combustion in a limited volume of the combustor [45]. Furthermore, at this scale the relative heat losses increase and may cause quenching of the reaction (fuel inside the combustion chamber that rapidly cools down, prevents it from burning) with consequent degradation of performances [13, 45, 65].Moreover, the engine speed and the gap width between the piston and the cylinder walls are two key parameters that can create issues in standard ICE (cylinder-piston engine)as reported by Sher et al. [58]. They simulated the miniaturization limits of a standard ICE with a rotation speed of 48000 rpm, a gap width of 10 um and a compression rate larger than 18 and they found a minimum size limit

#### 6 Energy Storage 118 Energy Storage – Technologies and Applications The Future of Energy Storage Systems <sup>7</sup>

between 0.3 and 0.4 cc. This limit has already been passed by [62] which has developed a microfabricated standard ICE of 5mmŒ3mmŒ1mm in dimension (0.015 cc) supplied by a mixture of Hydrogen and Oxygen, able to generate a mechanical power of 29.1mW. However, the tests of this silicon engine were performed at 3 rpm and it has shown a compression ratio around 1.4. A similar class of micro engines is based on rotary engine (Wankel design) instead of cylinder-piston design. Example of these MEMS engines can be found in [7, 29]. Interesting is the prototype developed by [7] which consisted of a 13 mm rotor diameter coupled with a dynamo meter, able to generate up to 4W of electric power (other versions of 90mW and 50W are under development). Another interesting alternative is using micro turbines. We are already used to find in the market large gas turbines for electric energy production that may generate up to several hundreds of Megawatts [51]. However, in the last years, with the advent of MNTs, many researchers started to investigate the possibility to create a small micro turbine that are able to generate few Watts, enough to supply most of today's electronic devices. The basic concept is similar to large scale turbine which consists on an upstream rotating compressor, a combustion chamber and a downstream turbine. Furthermore, since it will be used for generating electric power, there will also be an alternator coupled with the turbine. The fabrication material is mainly Silicon, thanks to the well established and controlled processing technologies available for this material. Furthermore, when compared with common nickel alloy, single crystal silicon has an *"'higher specific strength, it is quite oxidation-resistant and has thermal conductivity approaching that of copper, so it is resistant to thermal shock"'* [11]. Several authors [1, 33, 49] successfully designed micro turbines by using silicon (Si) as material. However, silicon has some limitation on high temperatures and for these reasons some other authors reported micro turbine fabricated whit other materials. In particular Peirs et al. [48] reported an example of a 36g micro turbine made of stainless steel. This 10mm diameter turbine was able to generate a maximum mechanical power of 28W with an efficiency of 18%. The turbine was then coupled to a small brushless dc motor, which was used as a three-phase generator. The total system was around 53mm long and 66g in weight, capable to generate 16W of electric power, corresponding to a total efficiency of 10.5%. A similar efficiency is also expected by another microturbine developed by Jacobson et al. [22] in which the preliminary test done so far are very encouraging. We should notice that even if the total efficiency is relative low, hydrogen and hydrocarbon fuels have a much higher energy density when compared with batteries (see fig.2), so the result is a substantial increase of the net energy density of the system.

**Table 1.** Super capacitors: Advantages and Drawbacks

& Ellenbogen [20], Kotz & Carlen [27]

defined as:

and drawbacks of supercapacitors respect to batteries are listed.

the maximum power will depend on the internal resistance (ESR) [55]. In TABLE 1 advantages

Figure4 is the Ragone plot for all three technologies. As we can see, supercapacitors will be able to fill the gap between standard capacitors and batteries. It should be noticed that despite supercapacitors there will certainly have a lower energy density of batteries, which

**Figure 4.** Ragone plot for Capacitors, Super Capacitors and Batteries; adapted from Everett [12], Halper

For better understanding the working principle of supercapacitors, it is convenient to start with some basic capacitor theory. The simplest capacitor (plane capacitor) consists of a two electrodes separated by a dielectric. Calling *d* the distance between the two electrodes, *S* their overlapped area and the dielectric constant of the dielectric, the total capacitance *C* can be

*C* = *S*/*d* (5)

The Future of Energy Storage Systems 119

can be easily recharged from any power network in seconds or fraction of seconds.

In conclusion, is certainly possible that ICE and MEMS gas turbines may one day be very useful as compact power sources for portable electronics, equipment, and small vehicles [11].

## **3. Supercapacitors**

As we briefly state above, batteries suffer from various limitation, such as limited life cycles, high manufacturing cost and relative low power density. Furthermore, in case of large batteries, they require also several hours for being fully charged. On the other hand, standard capacitors offer high power density, almost unlimited life cycles, and fast charge. However their energy density is currently too low for been used as primary energy storage system. Supercapacitors may combine the advantages of both battery and capacitor for creating a system that has high power density, virtual unlimited life cycles keeping at the same time acceptable energy density. In these devices, the internal leakage current (in the form of dipoles relax and/or charges re-combination) will determine how long the energy can be stored, while


**Table 1.** Super capacitors: Advantages and Drawbacks

6 Energy Storage

between 0.3 and 0.4 cc. This limit has already been passed by [62] which has developed a microfabricated standard ICE of 5mmŒ3mmŒ1mm in dimension (0.015 cc) supplied by a mixture of Hydrogen and Oxygen, able to generate a mechanical power of 29.1mW. However, the tests of this silicon engine were performed at 3 rpm and it has shown a compression ratio around 1.4. A similar class of micro engines is based on rotary engine (Wankel design) instead of cylinder-piston design. Example of these MEMS engines can be found in [7, 29]. Interesting is the prototype developed by [7] which consisted of a 13 mm rotor diameter coupled with a dynamo meter, able to generate up to 4W of electric power (other versions of 90mW and 50W are under development). Another interesting alternative is using micro turbines. We are already used to find in the market large gas turbines for electric energy production that may generate up to several hundreds of Megawatts [51]. However, in the last years, with the advent of MNTs, many researchers started to investigate the possibility to create a small micro turbine that are able to generate few Watts, enough to supply most of today's electronic devices. The basic concept is similar to large scale turbine which consists on an upstream rotating compressor, a combustion chamber and a downstream turbine. Furthermore, since it will be used for generating electric power, there will also be an alternator coupled with the turbine. The fabrication material is mainly Silicon, thanks to the well established and controlled processing technologies available for this material. Furthermore, when compared with common nickel alloy, single crystal silicon has an *"'higher specific strength, it is quite oxidation-resistant and has thermal conductivity approaching that of copper, so it is resistant to thermal shock"'* [11]. Several authors [1, 33, 49] successfully designed micro turbines by using silicon (Si) as material. However, silicon has some limitation on high temperatures and for these reasons some other authors reported micro turbine fabricated whit other materials. In particular Peirs et al. [48] reported an example of a 36g micro turbine made of stainless steel. This 10mm diameter turbine was able to generate a maximum mechanical power of 28W with an efficiency of 18%. The turbine was then coupled to a small brushless dc motor, which was used as a three-phase generator. The total system was around 53mm long and 66g in weight, capable to generate 16W of electric power, corresponding to a total efficiency of 10.5%. A similar efficiency is also expected by another microturbine developed by Jacobson et al. [22] in which the preliminary test done so far are very encouraging. We should notice that even if the total efficiency is relative low, hydrogen and hydrocarbon fuels have a much higher energy density when compared with batteries (see fig.2), so the result is a substantial increase

In conclusion, is certainly possible that ICE and MEMS gas turbines may one day be very useful as compact power sources for portable electronics, equipment, and small vehicles [11].

As we briefly state above, batteries suffer from various limitation, such as limited life cycles, high manufacturing cost and relative low power density. Furthermore, in case of large batteries, they require also several hours for being fully charged. On the other hand, standard capacitors offer high power density, almost unlimited life cycles, and fast charge. However their energy density is currently too low for been used as primary energy storage system. Supercapacitors may combine the advantages of both battery and capacitor for creating a system that has high power density, virtual unlimited life cycles keeping at the same time acceptable energy density. In these devices, the internal leakage current (in the form of dipoles relax and/or charges re-combination) will determine how long the energy can be stored, while

of the net energy density of the system.

**3. Supercapacitors**

the maximum power will depend on the internal resistance (ESR) [55]. In TABLE 1 advantages and drawbacks of supercapacitors respect to batteries are listed.

Figure4 is the Ragone plot for all three technologies. As we can see, supercapacitors will be able to fill the gap between standard capacitors and batteries. It should be noticed that despite supercapacitors there will certainly have a lower energy density of batteries, which can be easily recharged from any power network in seconds or fraction of seconds.

**Figure 4.** Ragone plot for Capacitors, Super Capacitors and Batteries; adapted from Everett [12], Halper & Ellenbogen [20], Kotz & Carlen [27]

For better understanding the working principle of supercapacitors, it is convenient to start with some basic capacitor theory. The simplest capacitor (plane capacitor) consists of a two electrodes separated by a dielectric. Calling *d* the distance between the two electrodes, *S* their overlapped area and the dielectric constant of the dielectric, the total capacitance *C* can be defined as:

$$\mathbf{C} = \varepsilon \mathbf{S} / d \tag{5}$$

#### 8 Energy Storage 120 Energy Storage – Technologies and Applications The Future of Energy Storage Systems <sup>9</sup>

which corresponds a stored energy E:

$$E = \frac{CV^2}{2} \tag{6}$$

**Figure 5.** Graphic representation (not in scale) of active carbon EDLC, adapted from [66, 70]

(ranging from 100 *F*/*g* to 300 *F*/*g*) than organic electrolytes (less than 150 *F*/*g*). [70].

calculate the roll up direction as shown in figure6.

availability and high cost, currently limits their usage [52].

despite the difficulties in preparation [70].

can be achieved [9, 20, 70].

penetrate dissolved in the nanopores. Furthermore it also concludes that the optimal pore size of the electrode mainly depends from the current load due to the distortion of cations and/or intercalation-like behavior [40]. This new discovery lead many scientists to re-consider roles of the micropores. The analytical models can be modified by splitting the capacitive behavior in two different parts depending on the pore size [40]. Despite of all these studies the EDLC is still not completely understood yet [59]. Active carbons super capacitors also change their capacitance respect to the electrolyte materials, aqueous electrolytes allow higher capacitances

The Future of Energy Storage Systems 121

For the carbon electrodes, one valid alternative to activated carbon is carbon nanotube (CNT), which consists of carbon atoms organized in cylindrical nanostructures and can be considered as rolled-up graphene sheets (which consist in carbons atoms organized in 2-D cells see fig.6). The roll up orientation is expressed by two indexes (*n* and *m*) and is very important in CNTs since different directions result different proprieties. The two indexes *n* and *m* are used for

Both SingleWalled (SWNTs) and Multiple Walled (MWNTs) were investigated for EDLC electrodes. Thanks to their high conductivity and their open shape both SWNTs and MWNTs are particularly suitable for high power density capacitors. Indeed their quickly accessible surface area and their easily tunable pore size enable electrolyte ions to diffuse into the mesopores (fig.7), therefore, reduced internal resistance (ESR) and increased maximum power

Unfortunately the specific surface area of CNT (< 500 *m*2/*g*) is much smaller than that of activated carbons ( 1,000U3,000 ˝ *m*2/*g*) [32, 70], resulting in lower energy density for the capacitor (in average between 1*Wh*/*kg* and 10*Wh*/*kg*)[52]. This, together with their limited

Beside Active Carbons and Carbon Nanotubes, in literature are presented many other materials that can be used for the EDCL electrodes. Among this we should cite carbon aerogel [54], (similar to gels but where the internal liquid is replaced with gas), xerogels [15] and carbon fibers [30]. Carbon aerogel electrode material gave promising capacitive properties,

where *V* is the voltage between the electrodes. From equations 5 and 6 it is clear that the energy density is proportional to the overlapped area and inversely proportional to the distance between the two electrodes. Thanks to nanotechnology it is possible to drastic reduce the effective distance d and create structures with large surface, resulting of an increasing of total capacitance. Moreover, it is possible to combine special types of dielectrics and ionic conducting liquid (electrolyte) in order to store not only electrostatic energy but also electrochemical energy. Supercapacitors can be classified into double-layer capacitor (EDLC) and electrochemical pseudo capacitor (EPC). Based on the storage mechanism we can divide supercapacitors into three categories [20]:


In the following sections we will give an overview of each class.

### **3.1. Electric double layer capacitors**

These supercapacitors are called also non-Faradaic supercapacitors since they do not involve any charge transfer between electrode and electrolyte. The energy storage mechanism is thus similar to standard capacitor where the area is much larger and the distance is in the atomic range of charges [70]. EDLC consist of two electrodes, one membrane between the two electrodes which separates the electrodes and electrolyte that can be either aqueous or non-aqueous depending on EDLC [23]. The material of the electrode is very important for the final supercapacitors performances. For the supercapacitors of today's innovation, the most common material of the electrodes is activated carbons because it is cheap, has large surface area and is easy to process [27, 70]. This material is organized in small hexagonal rings organized into graphene sheets [24]. The result is a large surface area due to the porous structure composed by micropores (< 2nm wide), mesopores (2 - 50 nm), and macropores (>50 nm) [20]. The basic structure of a carbon activated EDLC is shown in Fig.5.

For the analytic model of these capacitors equation 5 can still be considered true, where is the electrolyte dielectric constant, *S* is the specific surface area of the electrode accessible to the electrolyte ions, and *d* is the Debye length [70], however *"determination of the effective dielectric constant eff of the electrolyte and thickness of the double-layer formed at the interface is complex and not well understood"* [3]. Indeed we would expect that doubling the area of the active carbons would double also the capacitance. However experimental data are in contrast with the theory, since empiric measurements were showing a smaller capacitance than expected. Many scientist explained this phenomena by electrolyte ions that are too large to diffuse into smaller micropores and thus unable to support electrical double layer [20, 24, 69]. For this reason many authors have affirmed that mesopores are high desirable in EDLC electrodes since they can optimize their performances [8, 69]. However, recently [40] showed the important role of small pores in the EDCL and they affirmed that ions can

**Figure 5.** Graphic representation (not in scale) of active carbon EDLC, adapted from [66, 70]

8 Energy Storage

*<sup>E</sup>* <sup>=</sup> *CV*<sup>2</sup>

where *V* is the voltage between the electrodes. From equations 5 and 6 it is clear that the energy density is proportional to the overlapped area and inversely proportional to the distance between the two electrodes. Thanks to nanotechnology it is possible to drastic reduce the effective distance d and create structures with large surface, resulting of an increasing of total capacitance. Moreover, it is possible to combine special types of dielectrics and ionic conducting liquid (electrolyte) in order to store not only electrostatic energy but also electrochemical energy. Supercapacitors can be classified into double-layer capacitor (EDLC) and electrochemical pseudo capacitor (EPC). Based on the storage mechanism we can divide

These supercapacitors are called also non-Faradaic supercapacitors since they do not involve any charge transfer between electrode and electrolyte. The energy storage mechanism is thus similar to standard capacitor where the area is much larger and the distance is in the atomic range of charges [70]. EDLC consist of two electrodes, one membrane between the two electrodes which separates the electrodes and electrolyte that can be either aqueous or non-aqueous depending on EDLC [23]. The material of the electrode is very important for the final supercapacitors performances. For the supercapacitors of today's innovation, the most common material of the electrodes is activated carbons because it is cheap, has large surface area and is easy to process [27, 70]. This material is organized in small hexagonal rings organized into graphene sheets [24]. The result is a large surface area due to the porous structure composed by micropores (< 2nm wide), mesopores (2 - 50 nm), and macropores (>50

For the analytic model of these capacitors equation 5 can still be considered true, where is the electrolyte dielectric constant, *S* is the specific surface area of the electrode accessible to the electrolyte ions, and *d* is the Debye length [70], however *"determination of the effective dielectric constant eff of the electrolyte and thickness of the double-layer formed at the interface is complex and not well understood"* [3]. Indeed we would expect that doubling the area of the active carbons would double also the capacitance. However experimental data are in contrast with the theory, since empiric measurements were showing a smaller capacitance than expected. Many scientist explained this phenomena by electrolyte ions that are too large to diffuse into smaller micropores and thus unable to support electrical double layer [20, 24, 69]. For this reason many authors have affirmed that mesopores are high desirable in EDLC electrodes since they can optimize their performances [8, 69]. However, recently [40] showed the important role of small pores in the EDCL and they affirmed that ions can

<sup>2</sup> (6)

which corresponds a stored energy E:

supercapacitors into three categories [20]:

**3.1. Electric double layer capacitors**

3. Hybrid Supercapacitors

1. Electric Double Layer Capacitors (EDLC) 2. Electrochemical pseudo capacitors (EPC)

In the following sections we will give an overview of each class.

nm) [20]. The basic structure of a carbon activated EDLC is shown in Fig.5.

penetrate dissolved in the nanopores. Furthermore it also concludes that the optimal pore size of the electrode mainly depends from the current load due to the distortion of cations and/or intercalation-like behavior [40]. This new discovery lead many scientists to re-consider roles of the micropores. The analytical models can be modified by splitting the capacitive behavior in two different parts depending on the pore size [40]. Despite of all these studies the EDLC is still not completely understood yet [59]. Active carbons super capacitors also change their capacitance respect to the electrolyte materials, aqueous electrolytes allow higher capacitances (ranging from 100 *F*/*g* to 300 *F*/*g*) than organic electrolytes (less than 150 *F*/*g*). [70].

For the carbon electrodes, one valid alternative to activated carbon is carbon nanotube (CNT), which consists of carbon atoms organized in cylindrical nanostructures and can be considered as rolled-up graphene sheets (which consist in carbons atoms organized in 2-D cells see fig.6). The roll up orientation is expressed by two indexes (*n* and *m*) and is very important in CNTs since different directions result different proprieties. The two indexes *n* and *m* are used for calculate the roll up direction as shown in figure6.

Both SingleWalled (SWNTs) and Multiple Walled (MWNTs) were investigated for EDLC electrodes. Thanks to their high conductivity and their open shape both SWNTs and MWNTs are particularly suitable for high power density capacitors. Indeed their quickly accessible surface area and their easily tunable pore size enable electrolyte ions to diffuse into the mesopores (fig.7), therefore, reduced internal resistance (ESR) and increased maximum power can be achieved [9, 20, 70].

Unfortunately the specific surface area of CNT (< 500 *m*2/*g*) is much smaller than that of activated carbons ( 1,000U3,000 ˝ *m*2/*g*) [32, 70], resulting in lower energy density for the capacitor (in average between 1*Wh*/*kg* and 10*Wh*/*kg*)[52]. This, together with their limited availability and high cost, currently limits their usage [52].

Beside Active Carbons and Carbon Nanotubes, in literature are presented many other materials that can be used for the EDCL electrodes. Among this we should cite carbon aerogel [54], (similar to gels but where the internal liquid is replaced with gas), xerogels [15] and carbon fibers [30]. Carbon aerogel electrode material gave promising capacitive properties, despite the difficulties in preparation [70].

#### 10 Energy Storage 122 Energy Storage – Technologies and Applications The Future of Energy Storage Systems <sup>11</sup>

thus the capacitance will weakly depend on the surface materials of the electrodes [3]. Since these processes are more battery-like rather than capacitor-like, thus, the capacitors are named as the electrochemical pseudo supercapacitors (EPC). When compared with EDLC, EPC have smaller power density since Faradic processes are normally slower than that of non Faradic reactions. However EPCs can reach much higher capacitances [3, 59] and thus they store much

Currently research efforts are focused on investigating two types of materials for achieving large pseudo-capacitance: metal oxides and conductive polymers. Among all the metal oxide present in nature, Ruthenium oxide (*RuO*2) has been largely investigated [26, 31, 47, 63], thanks to its intrinsic reversibility for various surface redox couples and high conductivity [70]. In particular, the research has focused to explore the chemical reactions of *RuO*<sup>2</sup> in acid electrolytes. The results have explained the pseudo capacitance of EPC as adsorption of protons at Ruthenium Oxide surface, combined with a quick and reversible electron transfer,

where 0 *< x <* 2. With this type of electrode, the specific capacitance over 700*F*/*g* [31]. However, due to rarity and high cost, the commercial applications of *RuO*<sup>2</sup> supercapacitors have been limited. For this reason many researcher started to investigate others oxides that may provide the same performances with a lower costs. Manganese oxide (*MgOx*) has been an interesting candidate because of its low cost, nontoxic and large theoretical maximum capacitance about 1300*F*/*g*.. However the poor electronic and ionic conductivity, low surface area and difficulty to achieve long term cycling stability are some of the issue that need to be addressed before to make Manganese oxide usable in practice [31, 66]. Other metal oxides including (but not limited) *NiO*, *Ni*(*OH*)2, *Co*2*O*3, *IrO*2, *FeO*, *TiO*2, *SnO*2, *V*2, *O*<sup>5</sup> and *MoO*

Utilization of conducting polymers as the electrodes (polymer EPC) is also investigated. Since the polymer EPCs are based on a bulk process, a higher specific capacitance respect to carbon based capacitors can be achieved, and thus an expected larger energy density [36]. Furthermore conducting polymers are more conductive than the inorganic materials, thus, the reduced ESR and consequently the increased power capability respect to standard battery can be obtained [60]. Conductive polymers, as their name suggest are organic polymers that conduct electricity, are able to combine both advantages of metals (such as high conductivity) polymers (such as low cost, flexibility, low weight). They are made by doping conjugated polymers [21]. The synthesis of conjugated polymers can be done by chemical oxidation of the monomer or electrochemical oxidation of the monomer [21]. Polymer EPCs have three

basic configurations depending on the type of the polymer used [60]:

electro-activity (asymmetric configuration).

• Type I: Both electrodes use the same p-doped polymer (symmetric configuration).

and the n-doped for the negative electrode (symmetric configuration)

• Type II: Electrodes use two different p-doped polymers with a different range of

• Type III: Both electrodes use the same polymer ,with the p-doped for the positive electrode

Among the three configurations, TYPE III is the most promising one for commercialization [36, 60]. Despite it is possible to synthesize many types of conductive polymers, three in

*RuO*<sup>2</sup> <sup>+</sup> *xH*<sup>+</sup> <sup>+</sup> *xe*<sup>−</sup> <sup>↔</sup> *RuO*2−*x*(*OH*)*<sup>x</sup>* (7)

The Future of Energy Storage Systems 123

higher energy density.

as in 7 [59, 66]:

can also be applied [23].

**Figure 6.** Carbon nanotube: orientation indexes: adapted from GNU license images

**Figure 7.** Graphic representation (not in scale) of carbon nanotube EDLC, adapted from [24]

#### **3.2. Electrochemical pseudo supercapacitors**

Unlike EDLC, Electrochemical pseudo supercapacitors use charge transfer between electrode and electrolyte for storing energy. This Faradic process is achieved mainly by [20, 66, 70]:


The first two processes belong to surface mechanism, so the capacitance will strongly depend on the surface materials of the electrodes, while the third one is more bulk-based process and thus the capacitance will weakly depend on the surface materials of the electrodes [3]. Since these processes are more battery-like rather than capacitor-like, thus, the capacitors are named as the electrochemical pseudo supercapacitors (EPC). When compared with EDLC, EPC have smaller power density since Faradic processes are normally slower than that of non Faradic reactions. However EPCs can reach much higher capacitances [3, 59] and thus they store much higher energy density.

10 Energy Storage

**Figure 6.** Carbon nanotube: orientation indexes: adapted from GNU license images

**Figure 7.** Graphic representation (not in scale) of carbon nanotube EDLC, adapted from [24]

• Rapid and reversible Red-Ox reactions between the electrodes and the electrolyte

• Doping and undoping of active conducting polymer material of the electrode

Unlike EDLC, Electrochemical pseudo supercapacitors use charge transfer between electrode and electrolyte for storing energy. This Faradic process is achieved mainly by [20, 66, 70]:

The first two processes belong to surface mechanism, so the capacitance will strongly depend on the surface materials of the electrodes, while the third one is more bulk-based process and

**3.2. Electrochemical pseudo supercapacitors**

• Surface adsorption of ions from the electrolyte

Currently research efforts are focused on investigating two types of materials for achieving large pseudo-capacitance: metal oxides and conductive polymers. Among all the metal oxide present in nature, Ruthenium oxide (*RuO*2) has been largely investigated [26, 31, 47, 63], thanks to its intrinsic reversibility for various surface redox couples and high conductivity [70]. In particular, the research has focused to explore the chemical reactions of *RuO*<sup>2</sup> in acid electrolytes. The results have explained the pseudo capacitance of EPC as adsorption of protons at Ruthenium Oxide surface, combined with a quick and reversible electron transfer, as in 7 [59, 66]:

$$\rm RuO\_2 + xH^+ + xe^- \leftrightarrow RuO\_{2-x}(OH)\_x \tag{7}$$

where 0 *< x <* 2. With this type of electrode, the specific capacitance over 700*F*/*g* [31]. However, due to rarity and high cost, the commercial applications of *RuO*<sup>2</sup> supercapacitors have been limited. For this reason many researcher started to investigate others oxides that may provide the same performances with a lower costs. Manganese oxide (*MgOx*) has been an interesting candidate because of its low cost, nontoxic and large theoretical maximum capacitance about 1300*F*/*g*.. However the poor electronic and ionic conductivity, low surface area and difficulty to achieve long term cycling stability are some of the issue that need to be addressed before to make Manganese oxide usable in practice [31, 66]. Other metal oxides including (but not limited) *NiO*, *Ni*(*OH*)2, *Co*2*O*3, *IrO*2, *FeO*, *TiO*2, *SnO*2, *V*2, *O*<sup>5</sup> and *MoO* can also be applied [23].

Utilization of conducting polymers as the electrodes (polymer EPC) is also investigated. Since the polymer EPCs are based on a bulk process, a higher specific capacitance respect to carbon based capacitors can be achieved, and thus an expected larger energy density [36]. Furthermore conducting polymers are more conductive than the inorganic materials, thus, the reduced ESR and consequently the increased power capability respect to standard battery can be obtained [60]. Conductive polymers, as their name suggest are organic polymers that conduct electricity, are able to combine both advantages of metals (such as high conductivity) polymers (such as low cost, flexibility, low weight). They are made by doping conjugated polymers [21]. The synthesis of conjugated polymers can be done by chemical oxidation of the monomer or electrochemical oxidation of the monomer [21]. Polymer EPCs have three basic configurations depending on the type of the polymer used [60]:


Among the three configurations, TYPE III is the most promising one for commercialization [36, 60]. Despite it is possible to synthesize many types of conductive polymers, three in particular are commonly used for polymer EPCs as listed below (the relative structures are shown in fig.8) [60, 66] :

Wang et al. [67] has reported one interesting example of this class of supercapacitors. The electrodes were made by graphene/*RuO*<sup>2</sup> and graphene/*Ni*(*OH*)2. The device has shown an energy density ≈ 48*Wh*/*kg* at a power density of ≈ 0.23*kW*/*kg*. Furthermore a high cycling stability has also been achieved. After 5000 cycles of charging and discharging at a current density of 10 A/g it has shown 8% reduction of the original capacitance. Another example of using carbon electrodes coated by conductive polymers was reported by An et al. [2]. They

The Future of Energy Storage Systems 125

For the Asymmetric supercapacitors, carbon material for one electrode and metal oxide or polymer material as second electrode are applied. One example of this technique is reported

The battery type supercapacitors are the most interesting candidates for the relative high energy density. Similar to the asymmetric, in this type of supercapacitors one electrode is made of carbon material, and the second one is made of a typical battery electrode material (such as Lithium). In particular one lithium- intercalated compounds (Nanostructured *Li*4*Ti*5*O*12 also known as LTO)has been extensively studied [4, 10, 38, 41, 42, 46], which can enable the cycling stability during Li intercalation/ deintercalation processes [38]. When coupled with carbon electrode, the charge storage can be realized by the mechanisms of a Li-ion battery at the negative electrode and a supercapacitor at the positive electrode [10]. The energy density for battery like electrode is very high compared to capacitors. Naoi et al. [41] reported an energy density for as high as 55 Wh/Kg. However the power density for the battery type supercapacitor is in general lower than the other classes and, faradic principle leads to an increase in the energy density at the cost of life cycle. This is one major drawback of hybrid devices (when compared with EDLCs), and *"'it is important to avoid transforming a*

Tracking a clear path for the future energy storage systems is not easy. Fuel cells in the near future can became mature enough to be used as primary energy source for large vehicles. However, despite small fuel cells (in particular the direct methanol fuel cells) can also be used for supply laptop or other small electronic device, it is believed that battery will continue to lead this market for several years. Microturbines and internal combustion engines are also very promising technology but further research is needed for improving current prototypes. Supercapacitors have a great potential to play an important role in future energy storage systems, in particular to all the application that require high peek powers. Nowadays then can be used as secondary energy storage system, for example, in vehicle applications they can be used in parallel with fuel cells or batteries for overcoming the power peaks during acceleration [4]. Supercapacitors will be able to be full recharged in a very little time and this

provides the key advantage that the customers and the market are waiting for.

used polypyrrole-carbon aerogel for the capacitor electrodes.

*good supercapacitor into a mediocre battery"'* [59]

Luca Petricca, Per Ohlckers and Xuyuan Chen

*Vestfold University College, Norway*

**4. Conclusions**

**Author details**

by Staiti & Lufrano [61], which uses manganese oxide and activated carbon.


PANI has shown a high specific capacitance (around 800 *F*/*g* [66]) and a good life cycle stability (≈ −20% in capacitance after 9000 cycles) [60]. Polymer EPCs based on PPy and PTh are also successfully fabricated [36, 60, 66] with a specific capacitances between 200 and 300 *F*/*g*.

**Figure 8.** Polymers structures: polyaniline PANI(a), polypyrrole PPy(b), polythiophene PTh (c); adapted from [60]

In general, polymer EPCs suffer of poor cycle stability. The rapid performance degradation is due to the mechanical stress derived from the volumetric changes during the doping/dedoping process. (Swelling and shrinking)[16, 20, 27, 66]. For this reason further research is needed in order to improve performance of polymer EPCs.

## **3.3. Hybrid supercapacitors**

Hybrids supercapacitors are a class of devices that attempt to combine the advantages of both EDLC and EPC, which are able to exploit at the same time both Faradic and non- Faradic processes for storing energy. The idea is using the propriety of EDCL for obtaining high power and the propriety of EPC for increasing the energy density [59]. According to Halper & Ellenbogen [20], the hybrid supercapacitors can be classified, based on the electrode material, in:


In composite supercapacitors, each electrode is formed by a combination of a carbon material as the frame and a metal oxide material or a conductive polymer deposited on top of the carbon material (e.g. [67, 68]). BBy doing this the carbon material create a large surface area for a large capacitance. Furthermore the polymer material further increases the capacitance with a result of high energy density and cycling stability comparable with EDCL [20]. For example, in the case of Carbon/PPy based electrode, the high cycling stability is due to the carbon frame below the polymer that mitigates the polymer stress for increasing life cycle. Very recently Wang et al. [67] has reported one interesting example of this class of supercapacitors. The electrodes were made by graphene/*RuO*<sup>2</sup> and graphene/*Ni*(*OH*)2. The device has shown an energy density ≈ 48*Wh*/*kg* at a power density of ≈ 0.23*kW*/*kg*. Furthermore a high cycling stability has also been achieved. After 5000 cycles of charging and discharging at a current density of 10 A/g it has shown 8% reduction of the original capacitance. Another example of using carbon electrodes coated by conductive polymers was reported by An et al. [2]. They used polypyrrole-carbon aerogel for the capacitor electrodes.

For the Asymmetric supercapacitors, carbon material for one electrode and metal oxide or polymer material as second electrode are applied. One example of this technique is reported by Staiti & Lufrano [61], which uses manganese oxide and activated carbon.

The battery type supercapacitors are the most interesting candidates for the relative high energy density. Similar to the asymmetric, in this type of supercapacitors one electrode is made of carbon material, and the second one is made of a typical battery electrode material (such as Lithium). In particular one lithium- intercalated compounds (Nanostructured *Li*4*Ti*5*O*12 also known as LTO)has been extensively studied [4, 10, 38, 41, 42, 46], which can enable the cycling stability during Li intercalation/ deintercalation processes [38]. When coupled with carbon electrode, the charge storage can be realized by the mechanisms of a Li-ion battery at the negative electrode and a supercapacitor at the positive electrode [10]. The energy density for battery like electrode is very high compared to capacitors. Naoi et al. [41] reported an energy density for as high as 55 Wh/Kg. However the power density for the battery type supercapacitor is in general lower than the other classes and, faradic principle leads to an increase in the energy density at the cost of life cycle. This is one major drawback of hybrid devices (when compared with EDLCs), and *"'it is important to avoid transforming a good supercapacitor into a mediocre battery"'* [59]

## **4. Conclusions**

12 Energy Storage

particular are commonly used for polymer EPCs as listed below (the relative structures are

PANI has shown a high specific capacitance (around 800 *F*/*g* [66]) and a good life cycle stability (≈ −20% in capacitance after 9000 cycles) [60]. Polymer EPCs based on PPy and PTh are also successfully fabricated [36, 60, 66] with a specific capacitances between 200 and

**Figure 8.** Polymers structures: polyaniline PANI(a), polypyrrole PPy(b), polythiophene PTh (c); adapted

In general, polymer EPCs suffer of poor cycle stability. The rapid performance degradation is due to the mechanical stress derived from the volumetric changes during the doping/dedoping process. (Swelling and shrinking)[16, 20, 27, 66]. For this reason further

Hybrids supercapacitors are a class of devices that attempt to combine the advantages of both EDLC and EPC, which are able to exploit at the same time both Faradic and non- Faradic processes for storing energy. The idea is using the propriety of EDCL for obtaining high power and the propriety of EPC for increasing the energy density [59]. According to Halper & Ellenbogen [20], the hybrid supercapacitors can be classified, based on the electrode material,

In composite supercapacitors, each electrode is formed by a combination of a carbon material as the frame and a metal oxide material or a conductive polymer deposited on top of the carbon material (e.g. [67, 68]). BBy doing this the carbon material create a large surface area for a large capacitance. Furthermore the polymer material further increases the capacitance with a result of high energy density and cycling stability comparable with EDCL [20]. For example, in the case of Carbon/PPy based electrode, the high cycling stability is due to the carbon frame below the polymer that mitigates the polymer stress for increasing life cycle. Very recently

research is needed in order to improve performance of polymer EPCs.

shown in fig.8) [60, 66] :

• Polyaniline (PANI) • Polypyrrole (PPy)

300 *F*/*g*.

from [60]

in:

1. Composite 2. Asymmetric 3. Battery type

• Thiophene-based polymers (PTh)

**3.3. Hybrid supercapacitors**

Tracking a clear path for the future energy storage systems is not easy. Fuel cells in the near future can became mature enough to be used as primary energy source for large vehicles. However, despite small fuel cells (in particular the direct methanol fuel cells) can also be used for supply laptop or other small electronic device, it is believed that battery will continue to lead this market for several years. Microturbines and internal combustion engines are also very promising technology but further research is needed for improving current prototypes. Supercapacitors have a great potential to play an important role in future energy storage systems, in particular to all the application that require high peek powers. Nowadays then can be used as secondary energy storage system, for example, in vehicle applications they can be used in parallel with fuel cells or batteries for overcoming the power peaks during acceleration [4]. Supercapacitors will be able to be full recharged in a very little time and this provides the key advantage that the customers and the market are waiting for.

## **Author details**

Luca Petricca, Per Ohlckers and Xuyuan Chen *Vestfold University College, Norway*

## **5. References**

[1] Amit Mehra, I. A. W. & Schmidt, M. A. [1999]. Combustion tests in the 6-wafer static structure of a micro gas turbine engine, *TRANSDUCERS '99*.

[16] Frackowiak, E., Khomenko, V., Jurewicz, K., Lota, K. & Béguin, F. [2006]. Supercapacitors based on conducting polymers/nanotubes composites, *Journal of Power Sources*

The Future of Energy Storage Systems 127

URL: *http://www.fuelcellstore.com/en/pc/viewPrd.asp?idproduct=182&idcategory=40* [18] Gu, J. F., Gorgutsa, S. & Skorobogatiy, M. [2010]. Soft capacitor fibers using conductive polymers for electronic textiles, *Smart Materials and Structures* 19(11): 115006–.

[19] Gurau, B. & Smotkin, E. S. [2002]. Methanol crossover in direct methanol fuel cells: a link

[20] Halper, M. S. & Ellenbogen, J. C. [2006]. Supercapacitors : A brief overview, *Group*

[21] Harun, M.H., S. E. K. A. Y. N. . M. E. [2007]. Conjugated conducting polymers: A brief overview, *UCSI Academic Journal: Journal for the Advancement of Science & Arts* 2: 63–68. [22] Jacobson, S., Das, S., Savoulides, N., Steyn, J., Lang, J., Li, H., Livermore, C., Schmidt, M., Teo, C., Umans, S., Epstein, A., Arnold, D. P., Park, J., Zana, I. & Allen, M. [2004]. Progress toward a microfabricated gas turbine generator for soldier portable power applications,

[23] Jayalakshmi, M. & Balasubramanian, K. [2008]. Simple capacitors to supercapacitors - an

[24] J.M. Boyea, R.E. Camacho, S. P. T. & Ready, W. J. [2007]. Carbon nanotube-based supercapacitors: Technologies and markets, *Nanotechnology Law and Business* 4: 585–593. [25] Jost, K., Perez, C. R., McDonough, J. K., Presser, V., Heon, M., Dion, G. & Gogotsi, Y. [2011]. Carbon coated textiles for flexible energy storage, *Energy Environ. Sci.*

[26] Kim, I.-H. & Kim, K.-B. [2001]. Ruthenium oxide thin film electrodes for supercapacitors,

[27] Kotz, R. & Carlen, M. [2000]. Principles and applications of electrochemical capacitors,

[28] Kuwata, N., Kawamura, J., Toribami, K., Hattori, T. & Sata, N. [2004]. Thin-film lithium-ion battery with amorphous solid electrolyte fabricated by pulsed laser

[29] Lee, C., Jiang, K., Jin, P. & Prewett, P. [2004]. Design and fabrication of a micro wankel engine using mems technology, *Microelectronic Engineering* 73–74(0): 529–534.

[30] Leitner, K., Lerf, A., Winter, M., Besenhard, J., Villar-Rodil, S., Suarez-Garcia, F., Martinez-Alonso, A. & Tascan, J. [2006]. Nomex-derived activated carbon fibers as electrode materials in carbon based supercapacitors, *Journal of Power Sources*

URL: *http://www.sciencedirect.com/science/article/pii/S0013468600003546*

URL: *http://www.sciencedirect.com/science/article/pii/S1388248104000347*

URL: *http://www.sciencedirect.com/science/article/pii/S0167931704002060*

URL: *http://www.sciencedirect.com/science/article/pii/S0378775305007238*

deposition, *Electrochemistry Communications* 6(4): 417–421.

between power and energy density, *Journal of Power Sources* 112: 339–352.

URL: *http://www.sciencedirect.com/science/article/pii/S0378775305007391*

URL: *http://stacks.iop.org/0964-1726/19/i=11/a=115006*

*Proc. 24th Army Science Conf.*, Orlando, FL.

URL: *http://dx.doi.org/10.1039/C1EE02421C*

*Electrochem. Solid-State Lett.* 4(5): A62–A64. URL: *http://dx.doi.org/10.1149/1.1359956*

*Electrochimica Acta* 45(15–16): 2483–2498.

overview, *Int. J. Electrochem. Sci* 3: 1196 – 1217.

153(2): 413–418.

[17] FuelCellStore [2012].

1(March): 1–41.

4(12): 5060–5067.

153(2): 419–423.

[2] An, H., Wang, Y., Wang, X., Zheng, L., Wang, X., Yi, L., Bai, L. & Zhang, X. [2010]. Polypyrrole/carbon aerogel composite materials for supercapacitor, *Journal of Power Sources* 195(19): 6964–6969.

URL: *http://www.sciencedirect.com/science/article/pii/S0378775310007093*

[3] Andrew, B. [2000]. Ultracapacitors: why, how, and where is the technology, *Journal of Power Sources* 91(1): 37–50.

URL: *http://www.sciencedirect.com/science/article/pii/S0378775300004857*


14 Energy Storage

[1] Amit Mehra, I. A. W. & Schmidt, M. A. [1999]. Combustion tests in the 6-wafer static

[2] An, H., Wang, Y., Wang, X., Zheng, L., Wang, X., Yi, L., Bai, L. & Zhang, X. [2010]. Polypyrrole/carbon aerogel composite materials for supercapacitor, *Journal of Power*

[3] Andrew, B. [2000]. Ultracapacitors: why, how, and where is the technology, *Journal of*

[4] Balducci, A., Bardi, U., Caporali, S., Mastragostino, M. & Soavi, F. [2004]. Ionic liquids

[5] Cheng, X., Shi, Z., Glass, N., Zhang, L., Zhang, J., Song, D., Liu, Z.-S., Wang, H. & Shen, J. [2007]. A review of pem hydrogen fuel cell contamination: Impacts, mechanisms, and

[6] D. S. Nguyen, E. Halvorsen, G. U. J. & Vogl, A. [2010]. Fabrication and characterization of a wideband mems energy harvester utilizing nonlinear springs, *J. Micromech. Microeng*

[7] D.C. Walther, A. P. [2003]. Mems rotary engine power system: Project overview and recent research results, *SEM Annual Conference & Exposition on Experimental and Applied*

[8] Ding Sheng Yuan, Jianghua Zeng, J. C. Y. L. [2009]. Highly ordered mesoporous carbon synthesized via in situ template for supercapacitors, *Int. J. Electrochem. Sci.* 4: 562–570. [9] Du, C. & Pan, N. [2006]. High power density supercapacitor electrodes of carbon

[10] Du Pasquier, A., Plitz, I., Menocal, S. & Amatucci, G. [2003]. A comparative study of li-ion battery, supercapacitor and nonaqueous asymmetric hybrid devices for automotive

[11] Epstein, A. H. [2004]. Millimeter-scale, micro-electro-mechanical systems gas turbine

[12] Everett, M. A. [2008]. Ultracapacitors: A mainstream energy storage and power delivery

[13] Fernandez-Pello, A. C. [2002]. Micropower generation using combustion: Issues and

[14] Folkesson, A., Andersson, C., Alvfors, M. & Overgaard, L. [2003]. Real life testing of a

[15] Frackowiak, E. & Beguin, F. [2001]. Carbon materials for the electrochemical storage of

nanotube films by electrophoretic deposition, *Nanotechnology* 17(21): 5314–.

URL: *http://www.sciencedirect.com/science/article/pii/S0378775302007188*

approaches, *Proceedings of the Combustion Institute* 29(1): 883–899. URL: *http://www.sciencedirect.com/science/article/pii/S1540748902801134*

hybrid pem fuel cell bus, *Journal of Power Sources* 118(1–2): 349–357. URL: *http://www.sciencedirect.com/science/article/pii/S0378775303000867*

URL: *http://www.sciencedirect.com/science/article/pii/S0008622300001834*

structure of a micro gas turbine engine, *TRANSDUCERS '99*.

mitigation, *Journal of Power Sources* 165(2): 739–756.

URL: *http://stacks.iop.org/0957-4484/17/i=21/a=005*

applications, *Journal of Power Sources* 115(1): 171–178.

engines, *J. Eng. Gas Turbines Power* 126(2): 205–226.

URL: *http://dx.doi.org/10.1115/1.1739245*

energy in capacitors, *Carbon* 39(6): 937–950.

solution, KiloFarad International.

URL: *http://www.sciencedirect.com/science/article/pii/S0378775310007093*

URL: *http://www.sciencedirect.com/science/article/pii/S0378775300004857*

URL: *http://www.sciencedirect.com/science/article/pii/S0378775306025304*

for hybrid supercapacitors, *Electrochemistry Communications* 6(6): 566–570. URL: *http://www.sciencedirect.com/science/article/pii/S1388248104000803*

**5. References**

20: 125009.

*Mechanics*.

*Sources* 195(19): 6964–6969.

*Power Sources* 91(1): 37–50.

	- URL: *http://dx.doi.org/10.1039/C1EE02421C*

URL: *http://www.sciencedirect.com/science/article/pii/S0378775305007238*

	- [31] Lokhande, C., Dubal, D. & Joo, O.-S. [2011]. Metal oxide thin film based supercapacitors, *Current Applied Physics* 11(3): 255–270. URL: *http://www.sciencedirect.com/science/article/pii/S1567173910004773*

*Power Sources* 113(1): 62–71.

URL: *http://www.sciencedirect.com/science/article/pii/S0378775302004913*

URL: *http://www.sciencedirect.com/science/article/pii/S0169433208023234*

URL: *http://www.sciencedirect.com/science/article/pii/S0924424704000081*

generation, *Sensors and Actuators A: Physical* 113(1): 86–93.

*Proc. SPIE*, Vol. 7833, SPIE, Toulouse, France, pp. 783306–9.

*power-generation/gas-turbines/downloads/SGT5-8000H\_benefits.pdf*

[53] PowerStream, Lund Instrument Engineering, I. C. [2012]. URL: *http://www.powerstream.com/methanol-fuel-cell.htm*

carbon aerogels, *Journal of Power Sources* 105(2): 189–194.

URL: *http://dx.doi.org/10.1117/12.862729*

URL: *http://dx.doi.org/10.1155/2011/214549*

art.

GmbH.

7(11): 845–854.

*Power Sources* 166: 177–193.

URL: *http://dx.doi.org/10.1038/nmat2297*

[47] Patake, V., Lokhande, C. & Joo, O. S. [2009]. Electrodeposited ruthenium oxide thin films for supercapacitor: Effect of surface treatments, *Applied Surface Science* 255(7): 4192–4196.

The Future of Energy Storage Systems 129

[48] Peirs, J., Reynaerts, D. & Verplaetsen, F. [2004]. A microturbine for electric power

[49] Pelekies, S. O., Schuhmann, T., Gardner, W. G., Camacho, A. & Protz, J. M. [2010]. Fabrication of a mechanically aligned single-wafer mems turbine with turbocharger,

[50] Petricca, L., Ohlckers, P. & Grinde, C. [2011]. Micro- and nano-air vehicles: State of the

[51] Phil Ratliff, P. G. & Fischer, W. [2007]. *The New Siemens Gas Turbine SGT5-8000H for More Customer Benefit*, Siemens. URL: *http://www.energy.siemens.com/us/pool/hq/*

[52] Pint, C. L., Nicholas, N. W., Xu, S., Sun, Z., Tour, J. M., Schmidt, H. K., Gordon, R. G. & Hauge, R. H. [2011]. Three dimensional solid-state supercapacitors from aligned

[54] Probstle, H., Schmitt, C. & Fricke, J. [2002]. Button cell supercapacitors with monolithic

[56] Sepeur, S. [2008]. *Nanotechnology: technical basics and applications*, Vincentz Network

[57] Shah, K. & Besser, R. [2007]. Key issues in the microchemical systems-based methanol fuel processor: Energy density, thermal integration, and heat loss mechanisms, *Journal of*

[58] Sher, I., Levinzon-Sher, D. & Sher, E. [2009]. Miniaturization limitations of hcci internal

[59] Simon, P. & Gogotsi, Y. [2008]. Materials for electrochemical capacitors, *Nat Mater*

[60] Snook, G. A., Kao, P. & Best, A. S. [2011]. Conducting-polymer-based supercapacitor

[61] Staiti, P. & Lufrano, F. [2010]. Investigation of polymer electrolyte hybrid supercapacitor based on manganese oxide-carbon electrodes, *Electrochimica Acta* 55(25): 7436–7442.

[62] Suzuki, Y., Okada, Y., Ogawa, J., Sugiyama, S. & Toriyama, T. [2008]. Experimental study on mechanical power generation from mems internal combustion engine, *Sensors and*

single-walled carbon nanotube array templates, *Carbon* 49(14): 4890–4897. URL: *http://www.sciencedirect.com/science/article/pii/S0008622311005549*

URL: *http://www.sciencedirect.com/science/article/pii/S0378775301009387* [55] Reynolds, C. [2009]. Evolution of supercapacitors, AVX Application Engineer.

combustion engines, *Applied Thermal Engineering* 29(2–3): 400–411. URL: *http://www.sciencedirect.com/science/article/pii/S1359431108001427*

devices and electrodes, *Journal of Power Sources* 196(1): 1–12.

URL: *http://www.sciencedirect.com/science/article/pii/S0378775310010712*

URL: *http://www.sciencedirect.com/science/article/pii/S0013468610000824*


URL: *http://www.sciencedirect.com/science/article/pii/S0378775310000327*


URL: *http://www.sciencedirect.com/science/article/pii/S0378775312001851*


*Power Sources* 113(1): 62–71.

16 Energy Storage

[31] Lokhande, C., Dubal, D. & Joo, O.-S. [2011]. Metal oxide thin film based supercapacitors,

[32] Lu, W. & Dai, L. [2010-03-01 N1]. Carbon nanotube supercapacitors, *Carbon Nanotube*,

[33] Luc G. Frechette, Changgu Lee, S. A. & Liu, Y.-C. [2003]. Preliminary design of a mems steam turbine power plant-on-a-chip, *3rd IntŠl Workshop on Micro & Nano Tech. for Power*

[34] Lucintel [2011]. Growth opportunities in global battery market 2011-2016: Market size,

[35] Mandal, S., Bhojwani, P., Mohanty, S. & Mahapatra, R. [2008]. Intellbatt: Towards smarter battery design, *Design Automation Conference, 2008. DAC 2008. 45th ACM/IEEE*, pp. 872

[36] Mastragostino, M., Arbizzani, C. & Soavi, F. [2002]. Conducting polymers as electrode

[38] Mladenov, M., Alexandrova, K., Petrov, N., Tsyntsarski, B., Kovacheva, D., Saliyski, N. & Raicheff, R. [n.d.]. Synthesis and electrochemical properties of activated carbons and li<sub>4</sub>ti<sub>5</sub>o<sub>12</sub>

[39] Muller, D. C. [2007]. Energy storage: Strong momentum in high-end batteries. Credit

[40] Mysyk, R., Raymundo-Pinero, E. & Baguin, F. [2009]. Saturation of subnanometer pores in an electric double-layer capacitor, *Electrochemistry Communications* 11(3): 554–556.

[41] Naoi, K., Ishimoto, S., Isobe, Y. & Aoyagi, S. [2010]. High-rate nano-crystalline li4ti5o12 attached on carbon nano-fibers for hybrid supercapacitors, *Journal of Power Sources*

[42] Ni, J., Yang, L., Wang, H. & Gao, L. [n.d.]. A high-performance hybrid supercapacitor with li<sub>4</sub>ti<sub>5</sub>o<sub>12</sub>-c

[43] Odile BERTOLDI, S. B. [2009]. Report on energy, *Technical report*, European

[44] Ogawa, M., Kanda, R., Yoshida, K., Uemura, T. & Harada, K. [2012]. High-capacity thin film lithium batteries with sulfide solid electrolytes, *Journal of Power Sources*

[45] Park, D.-E., Lee, D.-H., Yoon, J.-B., Kwon, S. & Yoon, E. [2002]. Design and fabrication of micromachined internal combustion engine as a power source for microsystems, *Micro Electro Mechanical Systems, 2002. The Fifteenth IEEE International Conference on*, pp. 272

[46] Pasquier, A. D., Plitz, I., Gural, J., Menocal, S. & Amatucci, G. [2003]. Characteristics and performance of 500 f asymmetric hybrid advanced supercapacitor prototypes, *Journal of*

URL: *http://www.sciencedirect.com/science/article/pii/S1388248108006310*

URL: *http://www.sciencedirect.com/science/article/pii/S0378775310000327*

nano-composite prepared by in situ and ex situ carbon modification.

URL: *http://www.sciencedirect.com/science/article/pii/S0378775312001851*

URL: *http://www.sciencedirect.com/science/article/pii/S1567173910004773*

*Current Applied Physics* 11(3): 255–270.

*Generation & Energy Conv. (PowerMEMSŠ03)*.

market share and forecast analysis. Lucintel's report.

[37] Mench, M. M. [2008]. *Fuel cell engines*, John Wiley and Sons.

as electrode materials for supercapacitors. URL: *http://dx.doi.org/10.1007/s10008-011-1424-6*

URL: *http://dx.doi.org/10.1007/s10008-012-1704-9*

Commission-Observatory nano.

materials in supercapacitors, *Solid State Ionics* 148(3–4): 493–498. URL: *http://www.sciencedirect.com/science/article/pii/S0167273802000930*

InTech, pp. –.

–877.

Suisse, Suisse.

195(18): 6250–6254.

205(0): 487–490.

–275.

URL: *http://www.sciencedirect.com/science/article/pii/S0378775302004913*


URL: *http://dx.doi.org/10.1155/2011/214549*

	- URL: *http://dx.doi.org/10.1038/nmat2297*

18 Energy Storage <sup>130</sup> Energy Storage – Technologies and Applications **Chapter 6** 

*Actuators A: Physical* 141(2): 654–661.

URL: *http://www.sciencedirect.com/science/article/pii/S0924424707006334*


URL: *http://www.sciencedirect.com/science/article/pii/S0360128510000754*


© 2013 Xiao et al., licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

© 2013 Xiao et al., licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

**Analysis and Control of** 

Yong Xiao, Xiaoyu Ge and Zhe Zheng

stiffness (Cimuca et al., 2006; Park et al., 2008).

nonlinear (Yang & Chang, 1996).

http://dx.doi.org/10.5772/52412

**1. Introduction** 

Additional information is available at the end of the chapter

**Flywheel Energy Storage Systems** 

Since a few years ago, electrical energy storage has been attractive as an effective use of electricity and coping with the momentary voltage drop. Above all, flywheel energy storage systems (FESS) using superconductor have advantages of long life, high energy density, and high efficiency (Subkhan & Komori, 2011), and is now considered as enabling technology for many applications, such as space satellites and hybrid electric vehicles (Samineni et al., 2006; Suvire & Mercado, 2012). Also, the contactless nature of magnetic bearings brings up low wear, absence of lubrication and mechanical maintenance, and wide range of work temperature (Bitterly, 1998; Beach & Christopher, 1998). Moreover, the closed-loop control of magnetic bearings enables active vibration suppression and on-line control of bearing

Active magnetic bearing is an open-loop unstable control problem. Therefore, an initial controller based on a rigid rotor model has to be introduced to levitate the rotor. In reality, the spinning rotor under the magnetic suspension may experience two kinds of whirl modes. The conical whirl mode gives rise to the gyroscopic forces to twist the rotor, thereby severely affecting stability of the rotor if not properly controlled (Okada et al, 1992; Williams et al., 1990). The translatory whirl mode constrains the rotor to synchronous motion in the radial direction so as to suppress the gyroscopic rotation, which has been extensively used in industry (Tomizuka et al, 1992; Tsao et al., 2000). The synchronization control has also been shown to be very capable in dealing with nonlinear uncertain models, and to be very effective in disturbance rejection for systems subject to synchronous motion. Until the advent of synchronization control, the prevalent use of the synchronization controller has been limited to stable mechanical systems and therefore is not readily applicable to magnetic systems which are unstable in nature and highly

URL: *http://dx.doi.org/10.1039/B813846J*
