*5.2.1 Fault outside micro grid*

If the fault is in zone Z1 or Z2 then main grid protection system will clear the fault. As per the requirements of IEEE Standard 1547-2003, the micro grid has to be islanded by opening the CB1. If there are inverter based DERs in the micro grid, then the fault current will be limited by them. If conventional over current relays

**97**

*Microgrid Protection Systems*

be opened in 70 ms [20].

of consumers get affected.

micro grid earlier.

without any selectivity problem

considered.

*5.2.2 Fault inside the micro grid*

*DOI: http://dx.doi.org/10.5772/intechopen.86431*

depending on the number and type of DERs.

Ikmin = ∑

are used for tripping CB1, then the fault current will not be sufficient to trip the breaker. By employing a directional over current relay at LV bus, protection can be ensured. Alternatively changes in frequency or voltage can also be taken as useful indicators for detection of Islanding to initiate the desired protective action. The current setting of this relay should be the cumulative weighted sum of fault current contribution by all the DERs present in the micro grid governed by (Eq. (1)). The weighting factor varies from 1.1 (for inverter based DG) to 5 (for synchronous DG)

> 1 n

Here the Ikmin is the required adaptive relay current setting, kDER and IrDER are the weighting factor and rated current of the DER [19]. Based on the permissible voltage sag considerations, if sensitive loads are present in micro grid, CB1 should

If the fault occurs on the LV feeder or the consumer end, i.e., Z3 or Z4 then the protective system should isolate the faulty section ensuring that minimum number

Here again the two cases of grid connected and islanded modes of operation must be considered. Also the presence of inverter based DERs and synchronous based DERs should be given due consideration. Following are the key points to be

• If there is a fault in Z3 or Z4 in grid connected mode, main grid will supply

• If a large synchronous DER is present, then the fault current seen by the relay will be smaller than the fault current without DER causing protection blinding in case of a fault in Z3. It may also lead to delay in tripping the breaker if inverse definite minimum time (IDMT) over current relays are employed for protection. It is due to the fact that the IDMT relay characteristic has inverse characteristic for low magnitude portion of the fault current against the

• A low power diesel generator has low inertia. If there is a delay in the tripping, it might lead to unwanted tripping of the synchronous DER if the power rating is low. To avoid this, a proper adaptive coordination among the relays is essential.

• In islanded mode, if there is fault on Z3 and if there are inverter based DGs, they will limit the fault current as described in the case of faults outside the

• In islanded mode, if there is a fault in Z4, it can be isolated by proper relay setting based on the possible fault current supplied by the inverter based DERs

In a nutshell, the major challenge in over current protection is the potential difference in the fault currents due to the presence of DERs in grid connected and islanded mode. This calls for adaptive schemes which demand expensive and

sufficient fault current and faulty section will be isolated.

definite time characteristic for higher fault currents.

kDER ∗ IrDER (1)

**Figure 2.** *Typical micro grid showing the zones.*

*Micro-Grids - Applications, Operation, Control and Protection*

protection zone (Z4) as shown in **Figure 2**.

*5.2.1 Fault outside micro grid*

In order to understand the proper functioning of the overcurrent protection, let us consider a simple structure of a micro grid shown in **Figure 1**. In general this can be divided into four zones namely MV feeder and busbar protection zone (Z1), transformer protection zone (Z2), LV feeder protection zone (Z3) and micro grid

Based on the location of fault with respect to DER, they can be classified into external (in Z1 and Z2) and internal faults (in Z3 and Z4). If the CB1 is open the micro grid is in islanded mode and if it is closed it is in grid connected mode [18].

If the fault is in zone Z1 or Z2 then main grid protection system will clear the fault. As per the requirements of IEEE Standard 1547-2003, the micro grid has to be islanded by opening the CB1. If there are inverter based DERs in the micro grid, then the fault current will be limited by them. If conventional over current relays

**96**

**Figure 2.**

*Typical micro grid showing the zones.*

are used for tripping CB1, then the fault current will not be sufficient to trip the breaker. By employing a directional over current relay at LV bus, protection can be ensured. Alternatively changes in frequency or voltage can also be taken as useful indicators for detection of Islanding to initiate the desired protective action. The current setting of this relay should be the cumulative weighted sum of fault current contribution by all the DERs present in the micro grid governed by (Eq. (1)). The weighting factor varies from 1.1 (for inverter based DG) to 5 (for synchronous DG) depending on the number and type of DERs.

$$\mathbf{I}\_{\rm kmin} = \sum\_{\rm T}^{\rm R} \mathbf{k}\_{\rm DER} \ast \mathbf{I}\_{\rm rDER} \tag{1}$$

Here the Ikmin is the required adaptive relay current setting, kDER and IrDER are the weighting factor and rated current of the DER [19]. Based on the permissible voltage sag considerations, if sensitive loads are present in micro grid, CB1 should be opened in 70 ms [20].

### *5.2.2 Fault inside the micro grid*

If the fault occurs on the LV feeder or the consumer end, i.e., Z3 or Z4 then the protective system should isolate the faulty section ensuring that minimum number of consumers get affected.

Here again the two cases of grid connected and islanded modes of operation must be considered. Also the presence of inverter based DERs and synchronous based DERs should be given due consideration. Following are the key points to be considered.


In a nutshell, the major challenge in over current protection is the potential difference in the fault currents due to the presence of DERs in grid connected and islanded mode. This calls for adaptive schemes which demand expensive and complex communication infrastructure. The decision of disconnecting/keep it connected/shut down the micro grid depends on several factors such as reliability, cost and the number of customers that get affected. [18] Lot of research is focused on the application of the adaptive over current protection which demands effective communication infrastructure and the IEDs.

#### **5.3 Distance protection**

Based on the challenges of relay settings and coordination of the over current relays due the large difference in the fault currents in grid connected and islanded mode, research has been diverted towards the application of distance protection to micro grid in which the tripping decision is based on the impedance seen by the relay and not on the current magnitude [21, 22]. The DER output may result in under reach and power drawn by the loads may cause over reach of the distance relays. By employing more number of distance relays, these issues can be addressed. The impedance seen by the distance relay gets affected by the fault current limiting nature of the inverter based DERs. In case of induction motor generator based DGs employing SCIM (squirrel cage induction motor), when the machine starts absorbing reactive power, the line current leads the voltage. It poses the over reach problem to the connected distance relay which measures it. In case of a DFIM (doubly fed induction motor) based DG, the power factor of the DG unit is controlled by the control system of DFIM during fault conditions. If an unbalanced fault occurs and the fault currents are not large, then the control system can easily maintain the power factor of DFIM. It may lead to protection problems similar to that encountered in case of an inverter based DG [3]. This hinders the application of distance relays for protection of micro grid.

#### **5.4 Differential protection**

Difference between the measurements made at different points located in a micro grid (preferably at the two ends of a feeder section) is considered as an actuating quantity for this type of protection. Employing symmetrical components (zero sequence) a differential protection applying the directional features of the difference current can be used in three different ways as shown below [23].

In the first method (shown in **Figure 3**), in order to protect the micro grid and main grid a master micro grid control center (MGCC) is used. Using MGCC it is possible to integrate all protective schemes. Based on the information received from monitoring relays it is expected to protect the main grid and micro grid. However, this method is found to be costly and unreliable to protect either micro grid or main grid alone due to the complex communication infrastructure and the associated data analysis to be carried out.

Second method (shown in **Figure 4**) logic employed only local controllers. Every relay communicates with its neighboring relay directly and monitors the current direction. In this there is no master control center. Whenever a reversal of current is sensed, the faulted section is isolated.

Third method (refer to **Figure 5**) is an improvised version of second method. Each feeder has two monitoring relays. In this method, magnitude of the fault current also is considered in addition to direction unlike the previous two methods. With this the problem of low magnitude fault currents can be handled successfully.

Out of the three methods, second method is more cost effective. In the first method there would be a time delay as the data analysis has to be completed before the protective action is initiated and hence it cannot serve the purpose of primary

**99**

loads etc.

the micro grid [24].

*Method-3 improvized version with local controllers.*

*Microgrid Protection Systems*

*Method 1 with microgrid control center.*

**Figure 3.**

**Figure 4.**

**Figure 5.**

*Method 2 with local controllers.*

*DOI: http://dx.doi.org/10.5772/intechopen.86431*

protection. In the third method, there is an addition of one more directional monitoring unit in each feeder making it expensive. In all these three methods the fault detection and clearance are reliable and only the faulted section is isolated causing minimum number of consumers to be affected. These schemes do not require any change in the configuration or in the relay settings for both modes of operation of the micro grid and are independent of the type and number of DERs connected to

As there are no zero sequence currents in case of a phase to phase fault, negative sequence components of currents are used for fault detection [25]. Using the positive sequence components also considering both amplitude and phase angle the differential protection system is discussed in [26]. However, if there is unbalance and negative and zero sequence currents flow is due to unbalance in the micro grid rather than a fault, these methods need to be examined more carefully. Challenges in this type of protection may be summarized as high cost, communication infrastructure, need for synchronized measurements, effect of unbalanced

*Microgrid Protection Systems DOI: http://dx.doi.org/10.5772/intechopen.86431*

*Micro-Grids - Applications, Operation, Control and Protection*

communication infrastructure and the IEDs.

**5.3 Distance protection**

relays for protection of micro grid.

**5.4 Differential protection**

data analysis to be carried out.

sensed, the faulted section is isolated.

complex communication infrastructure. The decision of disconnecting/keep it connected/shut down the micro grid depends on several factors such as reliability, cost and the number of customers that get affected. [18] Lot of research is focused on the application of the adaptive over current protection which demands effective

Based on the challenges of relay settings and coordination of the over current relays due the large difference in the fault currents in grid connected and islanded mode, research has been diverted towards the application of distance protection to micro grid in which the tripping decision is based on the impedance seen by the relay and not on the current magnitude [21, 22]. The DER output may result in under reach and power drawn by the loads may cause over reach of the distance relays. By employing more number of distance relays, these issues can be addressed. The impedance seen by the distance relay gets affected by the fault current limiting nature of the inverter based DERs. In case of induction motor generator based DGs employing SCIM (squirrel cage induction motor), when the machine starts absorbing reactive power, the line current leads the voltage. It poses the over reach problem to the connected distance relay which measures it. In case of a DFIM (doubly fed induction motor) based DG, the power factor of the DG unit is controlled by the control system of DFIM during fault conditions. If an unbalanced fault occurs and the fault currents are not large, then the control system can easily maintain the power factor of DFIM. It may lead to protection problems similar to that encountered in case of an inverter based DG [3]. This hinders the application of distance

Difference between the measurements made at different points located in a micro grid (preferably at the two ends of a feeder section) is considered as an actuating quantity for this type of protection. Employing symmetrical components (zero sequence) a differential protection applying the directional features of the difference current can be used in three different ways as shown below [23].

In the first method (shown in **Figure 3**), in order to protect the micro grid and main grid a master micro grid control center (MGCC) is used. Using MGCC it is possible to integrate all protective schemes. Based on the information received from monitoring relays it is expected to protect the main grid and micro grid. However, this method is found to be costly and unreliable to protect either micro grid or main grid alone due to the complex communication infrastructure and the associated

Second method (shown in **Figure 4**) logic employed only local controllers. Every relay communicates with its neighboring relay directly and monitors the current direction. In this there is no master control center. Whenever a reversal of current is

Third method (refer to **Figure 5**) is an improvised version of second method. Each feeder has two monitoring relays. In this method, magnitude of the fault current also is considered in addition to direction unlike the previous two methods. With this the problem of low magnitude fault currents can be handled

Out of the three methods, second method is more cost effective. In the first method there would be a time delay as the data analysis has to be completed before the protective action is initiated and hence it cannot serve the purpose of primary

**98**

successfully.

**Figure 3.** *Method 1 with microgrid control center.*

**Figure 4.** *Method 2 with local controllers.*

**Figure 5.** *Method-3 improvized version with local controllers.*

protection. In the third method, there is an addition of one more directional monitoring unit in each feeder making it expensive. In all these three methods the fault detection and clearance are reliable and only the faulted section is isolated causing minimum number of consumers to be affected. These schemes do not require any change in the configuration or in the relay settings for both modes of operation of the micro grid and are independent of the type and number of DERs connected to the micro grid [24].

As there are no zero sequence currents in case of a phase to phase fault, negative sequence components of currents are used for fault detection [25]. Using the positive sequence components also considering both amplitude and phase angle the differential protection system is discussed in [26]. However, if there is unbalance and negative and zero sequence currents flow is due to unbalance in the micro grid rather than a fault, these methods need to be examined more carefully. Challenges in this type of protection may be summarized as high cost, communication infrastructure, need for synchronized measurements, effect of unbalanced loads etc.

### **5.5 Voltage based methodologies**

Extensive research has been carried out on these methods initially at University of Bath [27]. In this method voltage is considered for the detection of fault and subsequently for isolation. There are two methods. One is transformation method and the other is harmonic method.

### *5.5.1 Transformation method*

In this method, the output voltage of DER is transformed in two steps. (i) transform voltages from abc to dq frame using Eqs. (2) and (3).

$$
\begin{bmatrix} V\_{ds} \\ V\_{qs} \\ V\_0 \end{bmatrix} = \frac{2}{3} \begin{bmatrix} 1 & -1/\_2 & 1/\_2 \\ 0 & -\sqrt{3}/2 & \sqrt{3}/2 \\ 1/2 & 1/2 & 1/2 \end{bmatrix} \begin{bmatrix} V\_a \\ V\_b \\ V\_c \end{bmatrix} \tag{2}
$$

(ii) From dq transform to dc values

(ii) From dç transform to dc values 
$$
\begin{bmatrix} V\_{dr} \\ V\_{qr} \end{bmatrix} = \begin{bmatrix} \cos \alpha t & -\sin \alpha t \\ \sin \alpha t & \cos \alpha t \end{bmatrix} \begin{bmatrix} V\_{ds} \\ V\_{qs} \end{bmatrix} \tag{3}
$$

Any fault condition will get reflected as a change in d-q values.

$$\mathbf{V}\_{\rm DTST} = \mathbf{V}\_{q\rm ref} \mathbf{-} \mathbf{V} \tag{4}$$

**101**

*Microgrid Protection Systems*

*DOI: http://dx.doi.org/10.5772/intechopen.86431*

**5.7 Methods of improving protection**

tion level of the DERs is high.

**5.8 Protocols and standards**

**5.9 Recommendations**

online or following a time sequence. Thorough study of all possible topological configurations is to be carried out offline prior to the operation. It also necessitates conducting power flow studies and carrying out short circuit analysis for each configuration that might occur. For adjusting the settings and characteristics, fast

Considerable changes in fault current magnitudes during the grid connected and islanded modes of operation calls for alternative measures to be taken to improve the protection. If it is possible to modify the fault current magnitude whenever there is a change of operating mode of the micro grid, the existing protective systems can be used with some changes without the need of replacing them. If the fault current can be modified suitably by deploying some additional components, it would be very useful. These may be used either to increase or decrease the fault current suitably to have correct protective action along with the coordination among different protective equipment used. Response of a synchronous DER is different from an inverter fed DER during fault conditions. In case of an inverter fed DER, fault current need to be increased and in case of synchronous DG it should be reduced. Usage of fault current limiters (FCL), employing an interfacing unit at the point of micro grid interconnection with main grid to avoid the fault feeding from main grid are some of the available options. These options demand huge investment and maintenance. They depend on the proper functioning of islanding detection methods employed. Fault current limiting poses challenges if the size and penetra-

Some of the relevant standards related to micro grid operation are listed here for reference. IEEE Standard 1547 series covers Standard for Interconnecting Distributed Resources with Electric Power Systems. Standard Conformance Test Procedures for Equipment Interconnecting Distributed Resources with Electric Power Systems are given by IEEE Std 1547.1, Guide for Monitoring, Information Exchange, and Control of Distributed Resources Interconnected with Electric Power Systems is presented in IEEE Std 1547.3. Guide for Design, Operation, and Integration of Distributed Resource Island Systems with Electric Power Systems is IEEE Std 1547.4 and recommended Practice for Interconnecting Distributed Resources with Electric Power Systems Distribution Secondary Network are presented in IEEE Std 1547.6 [30]. There are reports prepared by CIGRE Working group also for reference. WG C6.22: Micro grid Evolution Roadmap contains the definitions and nomenclature of micro grid, WG C6.24 explains Capacity of Distribution Feeders for Hosting DER Connection and Integration of DER [31].

Of all the protection methods discussed above, differential current relaying is the most suited protection system for micro grid. This will enable fault location and also clearing in minimal time. Either one can have pilot lines for connecting the relays differentially (normally the feeder length in Distribution systems is not high so the cost of pilot lines also will be low) or one can locate RTUs at the two ends of each feeder and they will communicate the current magnitude, phase and direction to a central station where the fault location and tripping decisions are taken. The current measured at each end of the feeder is applied to Directional Over current

and effective communication infrastructure should be in place [29].

By comparing with the reference value, it can be easily inferred which type of fault and it can be isolated [27]. Application of transformations is an involved process and becomes complex in certain faults detection. Even a small difference in the voltage drop in case of a short line, shows a considerable effect on protection. Network topology also plays a major role in the application of this method when large numbers of DERs are present.

#### *5.5.2 Harmonic method*

In this method, when a fault occurs the total harmonic distortion (THD) of the terminal voltage increases. By comparing the THD of the terminal voltage of the converter with a predefined reference value, the type of fault can be identified. In this method discrete Fourier transforms are employed to convert the phase voltages Va, Vb, Vc into frequency domain. By using proper communication channel between the relays, fault area can be located and isolated [28]. This is used as backup protection. A correct setting for the reference value of THD is often challenging.

#### **5.6 Adaptive protection**

In this type of protection, the protection strategy must be modified in line with the existing operating conditions in the micro grid. It is to be done online. To accomplish this, numerical directional O/C relays are a good choice. Existing conventional fuses, electro mechanical and static relays settings and characteristics cannot be changed online. It necessitates that the existing protection equipment be upgraded to meet the requirement. Complying to IEC 61850 and installation of IEDs (Intelligent Electronic Devices) at appropriate places can make the relays to be adaptive with the ability to adjust their settings and characteristics accordingly on receiving the signals

#### *Microgrid Protection Systems DOI: http://dx.doi.org/10.5772/intechopen.86431*

*Micro-Grids - Applications, Operation, Control and Protection*

Extensive research has been carried out on these methods initially at University of Bath [27]. In this method voltage is considered for the detection of fault and subsequently for isolation. There are two methods. One is transformation method

In this method, the output voltage of DER is transformed in two steps. (i)

1 −1 ⁄2 <sup>1</sup> ⁄2 0 −√ \_\_ 3⁄2 <sup>√</sup> \_\_ 3⁄2

<sup>1</sup> ⁄2 <sup>1</sup> ⁄2 <sup>1</sup> ⁄2 ][

cos ω*t* − sin ω*t* sin <sup>ω</sup>*<sup>t</sup>* cos <sup>ω</sup>*<sup>t</sup>* ][

*VDIST* = *Vqref* − *V* (4)

In this method, when a fault occurs the total harmonic distortion (THD) of the terminal voltage increases. By comparing the THD of the terminal voltage of the converter with a predefined reference value, the type of fault can be identified. In this method discrete Fourier transforms are employed to convert the phase voltages Va, Vb, Vc into frequency domain. By using proper communication channel between the relays, fault area can be located and isolated [28]. This is used as backup protection.

In this type of protection, the protection strategy must be modified in line with the existing operating conditions in the micro grid. It is to be done online. To accomplish this, numerical directional O/C relays are a good choice. Existing conventional fuses, electro mechanical and static relays settings and characteristics cannot be changed online. It necessitates that the existing protection equipment be upgraded to meet the requirement. Complying to IEC 61850 and installation of IEDs (Intelligent Electronic Devices) at appropriate places can make the relays to be adaptive with the ability to adjust their settings and characteristics accordingly on receiving the signals

By comparing with the reference value, it can be easily inferred which type of fault and it can be isolated [27]. Application of transformations is an involved process and becomes complex in certain faults detection. Even a small difference in the voltage drop in case of a short line, shows a considerable effect on protection. Network topology also plays a major role in the application of this method when

*Va Vb Vc*

*Vds*

] (2)

*Vqs*] (3)

transform voltages from abc to dq frame using Eqs. (2) and (3).

⎤ ⎥ ⎦ = \_2 3[

⎡ ⎢ ⎣

*Vdr Vqr*] <sup>=</sup> [

Any fault condition will get reflected as a change in d-q values.

A correct setting for the reference value of THD is often challenging.

(ii) From dq transform to dc values

[

large numbers of DERs are present.

*5.5.2 Harmonic method*

**5.6 Adaptive protection**

*Vds Vqs V*<sup>0</sup>

**5.5 Voltage based methodologies**

and the other is harmonic method.

*5.5.1 Transformation method*

**100**

online or following a time sequence. Thorough study of all possible topological configurations is to be carried out offline prior to the operation. It also necessitates conducting power flow studies and carrying out short circuit analysis for each configuration that might occur. For adjusting the settings and characteristics, fast and effective communication infrastructure should be in place [29].
