**2. The gas turbine cycle**

In the past, STs were the prevailing type of PPs used to satisfy the baseload. GTs were included in these PPs to satisfy peak load and daily operation at time of high power demand lasting few hours in summer months and as emergency system. GTs for peak-load applications operate for short periods, few hours, e.g., 2–500 h/y, with no concern to thermal efficiency, but fast loading and start reliability are the concerns. Emergency GT units have to reach full load in a very short time; and aeroderivative GTs were originally designed to be capable of producing full power from cold metal in 120 seconds. The GT is usually classified as heavy frame industrial and aeroderivative types. Aeroderivative gas turbines use advanced aircraft engine to provide flexible, lightweight, and compact GTs. Heavy frame-type GTs are usually slower in speed, narrower in operating speed range, heavier, larger, have higher air flow, and slower in start-up. Traditionally, preference has been to place heavy frame industrial units in easily accessible baseload applications.

Developments of GT in terms of high unit capacity (up to 400 MW; see Figure 3a), reliability, and efficiency (up to 38 %) extend their use to cover base-load. Many simple GT cycle power plants are operating in theGCC, e.g., 497MWRasAbuFontas (RAF)inQatarincluding 6GT×32 MW, 6×48 MW, and 2×9 MWof GT units.It has also 10×7 MIGD capacity multistage flash (MSF) desalting units with steam supplied from HRSG. The plant was commissioned in 1980. The desalting capacity was reported as 55 MIGD in 2010. The GT combined with ST forms GTCC of higher efficiency than either GT or ST cycle. Another diagram of GT is given in Figure 3b.

**Figure 3.** a: Siemens SGT5-8000H of 400 MW capacity [3] b: Cutaway diagram of a Westinghouse 501D5A gas turbine [4]

Simple GT cycle (Figures 4a–d) consists of a compressor, turbine, and generator usually mounted on single shaft and combustion chamber (cc). When started, the generator is usually operated as a motor to get sufficient rotor speed. Then, the GT is ignited, and power supply to the generator-motor is switched off. The GT accelerates until it reaches its nominal speed, and generator is synchronized and connected to the power grid. The GT is operated at constant speed to keep constant frequency at the generator output. The load changes are compensated by the adjustment of the input fuel flow to the combustor.

This article presents description of GTCC power plant units, detailing the GT, HRSG, ST, and their combining arrangements with DP to form CPDP. Methods of allocating the fuel supplied

In the past, STs were the prevailing type of PPs used to satisfy the baseload. GTs were included in these PPs to satisfy peak load and daily operation at time of high power demand lasting few hours in summer months and as emergency system. GTs for peak-load applications operate for short periods, few hours, e.g., 2–500 h/y, with no concern to thermal efficiency, but fast loading and start reliability are the concerns. Emergency GT units have to reach full load in a very short time; and aeroderivative GTs were originally designed to be capable of producing full power from cold metal in 120 seconds. The GT is usually classified as heavy frame industrial and aeroderivative types. Aeroderivative gas turbines use advanced aircraft engine to provide flexible, lightweight, and compact GTs. Heavy frame-type GTs are usually slower in speed, narrower in operating speed range, heavier, larger, have higher air flow, and slower in start-up. Traditionally, preference has been to place heavy frame industrial units in

Developments of GT in terms of high unit capacity (up to 400 MW; see Figure 3a), reliability, and efficiency (up to 38 %) extend their use to cover base-load. Many simple GT cycle power plants are operating in theGCC, e.g., 497MWRasAbuFontas (RAF)inQatarincluding 6GT×32 MW, 6×48 MW, and 2×9 MWof GT units.It has also 10×7 MIGD capacity multistage flash (MSF) desalting units with steam supplied from HRSG. The plant was commissioned in 1980. The desalting capacity was reported as 55 MIGD in 2010. The GT combined with ST forms GTCC of higher efficiency than either GT or ST cycle. Another diagram of GT is given in Figure 3b.

(a) (b)

**Figure 3.** a: Siemens SGT5-8000H of 400 MW capacity [3] b: Cutaway diagram of a Westinghouse 501D5A gas turbine

to the CPDP using GTCC between the EP and DW are presented.

**2. The gas turbine cycle**

130 Desalination Updates

easily accessible baseload applications.

[4]

Another arrangement (Figures 5a, b) is to put the compressor and portion of the turbine on one shaft while the other part of the turbine and the generator on another shaft. The compressor with the first part of the turbine is called the gas generator (GG); and the GG output is equal to the compressor-consumed power. The other part of the turbine and generator is called the free turbine, which produces the net GT power output and gives more flexibility. Figure 4b shows GE's LM2500 Base aeroderivative gas turbine package that has dual fuel (oil and gas) capability; fast load response; 16-stage axial-flow compressor; annular combustor; two-stage, high-pressure, and single-rotor gas turbine; and highly efficient six-stage power turbine.

**Figure 4.** a: Gas turbine open (Brayton) cycle with its operating variables [5] b: Single-shaft gas turbine unit compo‐ nents [5] c: Gas turbine cycle presentation on P-v diagram [5] d: Gas turbine cycle presentation on T-s diagram

(c) **Figure 5.** a: Two-shaft GT. b: Two-shaft GE LM2500 aeroderivative GT [6]

#### **2.1. Analysis of ideal gas turbine cycle**

The simple ideal gas turbine open cycle (see Figure 4c) known as Brayton cycle consists of four processes:

(d)


The exhausted hot gas is released from the turbine to the atmosphere, and fresh air is used to start or continue the cycle. In fact it is not a real cycle, but process 4-1 can be considered as an isobaric process of heat rejection to atmosphere. The cycle can be represented on both pressurespecific volume (P-v) and temperature-entropy (T-s) diagrams as shown in Figure 4d.

When the air is considered as ideal gas, the property relations for process 1-2 are

$$\begin{aligned} PV &= RT\_\prime & P\_1V\_1 &= RT\_{1\prime} & P\_2V\_2 &= RT\_{2\prime} \\ PV^k &= C\_\prime & P\_1V^k{}\_1 &= P\_2V^k{}\_2 = C \end{aligned}$$

where

$$k = \mathbb{C}\_p / \mathbb{C}\_{v'} $$

and then

$$\frac{P\_2 V\_2}{P\_1 V\_1} = \frac{P\_2}{P\_1} (\frac{P\_2}{P\_1})^{-\frac{1}{k}} = \frac{RT\_2}{RT\_1}$$

The required compressor-specific process (work per kg) is

$$\begin{aligned} \text{ev}\_{c,is}(1 \to 2) &= -\mathop{\rm l}\limits\_{1}^{2} \text{v}dp = \mathop{\rm l}\limits\_{1}^{2} \text{C}\frac{\text{C}}{P^{1/k}}dP\\ &= \frac{k}{1-k}(P\_{2}\upsilon\_{2} - P\_{1}\upsilon\_{1}) = -\frac{k}{k-1}P\_{1}\upsilon\_{1}(\frac{P\_{2}\upsilon\_{2}}{p\_{1}\upsilon\_{1}} - 1) = -\frac{k}{k-1}RT\_{1}(\frac{P\_{2}}{p\_{1}})^{\frac{k-1}{k}} - 1) \end{aligned}$$

This is negative work (work consumed by the compressor) and is equal also to

$$\left| w\_{c,is}(\mathbf{1} \to \mathbf{2s}) \right| = (h\_{2s} - h\_1) = c\_p (T\_{2s} - T\_1)$$

T2s and h2s are the absolute temperature and enthalpy at point 2 if the expansion is isentropic (q1-2 = 0).

Heat addition process from 2 to 3 in the combustion chamber is considered ideal with no pressure loss (isobaric), P2 = P3. The heat input qin between 2 and 3 is equal to the enthalpy increase: *qin* =*h*<sup>3</sup> −*h*2*<sup>s</sup>* =*Cp*(*T*<sup>3</sup> −*T*2*s*).

It is noticed here that T3 is the highest temperature in the cycle and is called the turbine inlet temperature (TIT). The amount of specific heat input per kg of air is also equal to

$$q\_f = \frac{m\_f}{m\_a} \text{(LHV)}\_f$$

where *m*<sup>f</sup> is the mass flow rate of the fuel input and LHV is the fuel low heating value (heat generated per kg of fuel, when the water vapor in the combusted gases is in vapor state.

It is also noticed here that w2-3 = 0.

(a) (c)

**2.1. Analysis of ideal gas turbine cycle**

processes:

132 Desalination Updates

where

and then

**Figure 5.** a: Two-shaft GT. b: Two-shaft GE LM2500 aeroderivative GT [6]

pressor, with P2/P1= rp, called pressure ratio.

The required compressor-specific process (work per kg) is

**3.** Isentropic expansion of the working fluid in GT turbine from 3 to 4.

When the air is considered as ideal gas, the property relations for process 1-2 are

The simple ideal gas turbine open cycle (see Figure 4c) known as Brayton cycle consists of four

**1.** Isentropic compression of ambient air (working fluid) from pressure P1 to P2 by a com‐

**2.** Heat transfer to the working fluid by mixing fuel with the compressed air and combusted in the cc from 2 to 3; usually P2 is assumed equal to P3 for ideal cycle, i.e., isobaric process.

The exhausted hot gas is released from the turbine to the atmosphere, and fresh air is used to start or continue the cycle. In fact it is not a real cycle, but process 4-1 can be considered as an isobaric process of heat rejection to atmosphere. The cycle can be represented on both pressurespecific volume (P-v) and temperature-entropy (T-s) diagrams as shown in Figure 4d.

11 22

/ , *p v kC C* =

<sup>1</sup> 22 2 2 <sup>2</sup> 11 1 1 1 ( ) *<sup>k</sup> P V P P RT P V P P RT* - = =

, , , *k kk*

*PV C P V P V C*

*PV RT P V RT P V RT*

== = = ==

11 1 22 2

(b) (d)

The property relations of isentropic expansion process in the turbine can be expressed as

$$\begin{aligned} PV &= RT\_\prime & P\_3V\_3 &= RT\_{3\prime} & P\_4V\_4 &= RT\_4\\ PV^k &= C\_\prime & P\_3V\_3^k &= P\_4V^k{}\_4 &= C \end{aligned}$$

The turbine isentropic work is expressed by

$$\begin{aligned} \, ^2w\_{t,is}(3 \to 4s) &= \int\_3^4 vdp = \int\_1^2 C \frac{C}{P^{1/k}} dP = \frac{k}{1-k} (P\_4 v\_4 - P\_3 v\_3) = -\frac{k}{k-1} P\_3 v\_3 (\frac{P\_3 v\_3}{p\_4 v\_4} - 1) \\\ &= \frac{k}{k-1} R T\_3 (\frac{P\_3}{p\_4})^{\frac{k-1}{k}} - 1) \end{aligned}$$

This work can also be expressed by enthalpy change as

$$\left| w\_{t,is} \left( \mathfrak{Z} \to \mathfrak{4s} \right) \right| = \left( h\_3 - h\_{4s} \right) = c\_p \left( T\_3 - T\_{4s1} \right)$$

Since part of the turbine work is used to drive the compressor, the net work output (wnet = wt - wc) is expressed as

$$\begin{aligned} \text{tr}\_{\text{net}}\left(\text{ideal}\right) &= \text{w}\_{t} - \text{w}\_{c} = (\text{h}\_{3} - \text{h}\_{4s}) - (\text{h}\_{2s} - \text{h}\_{1}) = \text{c}\_{p}\left(\text{T}\_{3} - \text{T}\_{4s}\right) - \text{c}\_{p}\left(\text{T}\_{2s} - \text{T}\_{1}\right) \\ \text{tr}\_{\text{net}} &= \text{w}\_{t} - \text{w}\_{c} = \text{c}\_{p}T\_{3}\left(1 - \frac{\text{T}\_{4s}}{\text{T}\_{3}}\right) - \text{c}\_{p}T\_{1}\left(\frac{\text{T}\_{2s}}{\text{T}\_{1}} - 1\right) \\ &= \text{c}\_{p}T\_{3}\left(1 - \frac{1}{rp^{\frac{k-1}{k}}}\right) - \text{c}\_{p}T\_{1}\left(rp^{\frac{k-1}{k}} - 1\right) \end{aligned}$$

It is noticed that the net heat (*q*in *– q*out) is equal to the net work, or

$$\begin{aligned} \oint \mathcal{O} \mathcal{Q} \mathcal{Q} &= \oint \mathcal{O} \mathcal{W} = q\_{in} - q\_{out} = \mathcal{w}\_{out} - \mathfrak{w}\_{in} \\ &= c\_p \left( T\_3 - T\_{4s} \right) - c\_p \left( T\_{2s} - T\_1 \right) = c\_p \left( T\_3 - T\_{2s} \right) - q\_{out} \\ & q\_{out} = c\_p \left( T\_{4s} - T\_1 \right) = q\_{4,1} = c\_p \cdot \left( T\_1 - T\_4 \right) \end{aligned}$$

The ideal cycle efficiency (net work/heat in) is expressed by

$$\eta = \frac{w\_{\rm net}}{q\_{\rm in}} = \frac{q\_{\rm in} - q\_{\rm out}}{q\_{\rm in}} = 1 - \frac{q\_{\rm out}}{q\_{\rm in}} = 1 - \frac{T\_{4s} - T\_1}{T\_3 - T\_{2s}} = 1 - \frac{T\_1(\frac{T\_{4s}}{T\_1} - 1\_1)}{T\_2(\frac{T\_3}{T\_{2s}} - 1)}$$

It is noticed that

$$\begin{aligned} \frac{T\_{2s}}{T\_1} &= \left(\frac{P\_2}{P\_1}\right)^{\frac{k-1}{k}}, & \frac{T\_3}{T\_{4s}} &= \left(\frac{P\_3}{P\_4}\right)^{\frac{k-1}{k}}, & P\_1 &= P\_{4s}, P\_2 = P\_{3s},\\ \frac{T\_{2s}}{T\_1} &= \frac{T\_3}{T\_{4s}}, & T\_{2s}T\_{4s} &= T\_1T\_{3s}, & \frac{T\_{2s}}{T\_3} &= \frac{T\_1}{T\_{4s}}. \end{aligned}$$

Then

Cogeneration Power-Desalting Plants Using Gas Turbine Combined Cycle http://dx.doi.org/10.5772/60209 135

$$\eta c = 1 - \frac{T\_1}{T\_2} = 1 - \frac{1}{r p^{\frac{k-1}{k}}}$$

where

This work can also be expressed by enthalpy change as

3 1 1


*c T c T rp rp*

=- - -

*p p k k*

<sup>1</sup> (1 ) ( 1)

It is noticed that the net heat (*q*in *– q*out) is equal to the net work, or

Ñ Ñ ò ò

The ideal cycle efficiency (net work/heat in) is expressed by

h

It is noticed that

Then

*net t c p p*

*T T w w w cT cT*

=-= - - -


134 Desalination Updates

, 3 4 3 41 (3 4 ) ( ) ( ) *w s h h cT T t is* ® =- = - *s p <sup>s</sup>*

Since part of the turbine work is used to drive the compressor, the net work output (wnet = wt

( ) ( )( ) ( ) ( )

=-= - - - = - - -

4 2

*net t c s s p s ps s s*

*w ideal w w h h h h c T T c T T*

(1 ) ( 1)

*T T*

3 1 1

34 2 1 32 4 1 4,1 1 4

*w qq q TT T q q q TT <sup>T</sup> <sup>T</sup>*

1 1


*k k s k k s s s s s*

(), (), ,

*T TP <sup>P</sup> P PP P*

, ,,

*s s*

2 3 2 1 2 4 13 1 4 3 4

= = =

*T T <sup>T</sup> <sup>T</sup> T T TT T T T T*


*p s ps p s out*

( )( )( ) ( ) ()

= - - -= - - = - = =× -

*cT T cT T cT T q*

*Q Wq q w w*

¶=¶ = - = -

*q cT T q c T T*

*in out out in*

11 1

3 1


*k k*

*out p s p*

*net in out out s in in in s*

2 2 3 3

11 4 4

*TP T P*

34 2 1 34 2 1

4 1 1

*s*

( 1)

( 1)

*s*

*T*

2 2

*<sup>T</sup> <sup>T</sup>*

4 1 1 3 2 3

1 42 3

$$rp = (\frac{P\_2}{P\_1}) = (\frac{P\_3}{P\_4})^\gamma$$

It is noticed that the efficiency depends also on T3 (TIT), pressure ratio rp, and k, the ratio of specific heats at constant pressure to that at constant volume and is equal to 1.4 for air, *k* = *Cp Cv* . The dimensionless work output can be expressed by

$$\frac{w\_{net}}{C\_p T\_1} = \frac{T\_3}{T\_1} (1 - \frac{T\_{4s}}{T\_3}) - (\frac{T\_2}{T\_1} - 1) \\ \prime \\ \frac{T\_{2s}}{T\_1} = \frac{T\_3}{T\_1} (1 - \frac{1}{rp^{\frac{k-1}{k}}}) - (rp^{\frac{k-1}{k}} - 1) \\ v$$

Process (2-3): isobaric heat supply, q2-3 = h3 - h2, *wmech* ,2,3 =0. (4)

Process (3-4): isentropic expansion, wt = cp(T3-T4s). (5)

State (4-1): isobaric heat release, *q*4,1 =*cp* ⋅ (*T*<sup>1</sup> −*T*4*<sup>s</sup>*). (9)

There are differences between the ideal Brayton cycle and real gas turbine cycle. In the real cycle, the following are included:


The losses in compressors are usually expressed through the following:


#### **2.2. The GT performance**

The simple cycle in an *h-s* diagram including losses is shown in Figure 6a.

**Figure 6.** a: Enthalpy-entropy (h-s) diagram for ideal and practical gas turbine cycle [7] b: Dependence of the thermal efficiency ηth of the cycle on the parameters rp, k, and θ for ηt,is= 0.88 and ηc,is = 0.86. Line 1 joins points of maximum efficiency for each curve [7] c: Dependence of the specific work of the cycle on the parameters π, κ, and θ for η<sup>T</sup> =0.88 and ηC =0.86. Line 1 joins points of maximum specific work for each curve [7] d: Thermal efficiency vs specific power for varying pressure ratios (10–26) and combustor outlet temperature (1,473–1,773 K) for a gas turbine [8]

$$\begin{aligned} \eta\_c \text{(cycle efficiency)} &= \left(h\_3 - h\_4\right) / \left(h\_3 - h\_2\right) \\ \eta\_{t,is} \text{(turbine isentropic efficiency)} &= \left(h\_3 - h\_4\right) / \left(h\_3 - h\_{4s}\right) \\ \eta\_c \text{(compressor isentropic efficiency)} &= \frac{\mathbf{w}\_{c,is}}{\mathbf{w}\_c} = \frac{h\_{2s} - h\_1}{h\_2 - h\_1} \\ &= \frac{k}{k-1} RT\_1 \left[ \left(\frac{P\_2}{P\_1}\right)^{\frac{k-1}{k}} - 1 \right] = \frac{k}{k-1} RT\_1 \left[ \left(\frac{P\_2}{P\_1}\right)^{\frac{k-1}{k}} - 1 \right] / \mathbf{C}\_p \left(\mathbf{T}\_2 - \mathbf{T}\_1\right) \end{aligned}$$

The losses in turbines are usually expressed by the turbine efficiency defined by ηt = actual work of expansion/ideal work of expansion.

Assuming that θ is the ratio of the turbine inlet temperature and compressor inlet temperature, which in this case is θ = *T*3*/T*1,

**2.2. The GT performance**

136 Desalination Updates

The simple cycle in an *h-s* diagram including losses is shown in Figure 6a.

(

b)

(

d)

(a)

(c)

*c t is*

*c*

h

ηt

**Figure 6.** a: Enthalpy-entropy (h-s) diagram for ideal and practical gas turbine cycle [7] b: Dependence of the thermal efficiency ηth of the cycle on the parameters rp, k, and θ for ηt,is= 0.88 and ηc,is = 0.86. Line 1 joins points of maximum efficiency for each curve [7] c: Dependence of the specific work of the cycle on the parameters π, κ, and θ for η<sup>T</sup> =0.88 and ηC =0.86. Line 1 joins points of maximum specific work for each curve [7] d: Thermal efficiency vs specific power

( ) ( )

3' 4 3 2

)

*tropic efficiency h h*

1 1

*h h compressor isen*

*turbine isentropic efficiency h h h h*

=- -

2 1 1

é ù ê ú ë û <sup>=</sup> <sup>=</sup> ê ú - - - ê ú ë û

The losses in turbines are usually expressed by the turbine efficiency defined by

( ) ( ) c,is 2 1 c 21

*s*

4


*s*

C T – T p2 1 ( )

2 1

) <sup>1</sup>

( 1/

*k*


*k*


w

= -

w

for varying pressure ratios (10–26) and combustor outlet temperature (1,473–1,773 K) for a gas turbine [8]

, 3' 4 3'

η( /

*cycle efficiency h h h h*

1 2

*k P <sup>k</sup> <sup>P</sup> RT hh k P*

*k*

é ù ê ú - -

*k*


( )

1

η( ) /

() 1 <sup>1</sup>

= actual work of expansion/ideal work of expansion.

*<sup>k</sup> <sup>P</sup> RT*

$$\mathfrak{m}\_c = \frac{\frac{T\_3}{T\_1} \mathfrak{m}\_{\mathfrak{t},is} \left( 1 - \frac{1}{rp^{\frac{k-1}{k}}} \right) - \frac{1}{\mathfrak{m}\_{\mathfrak{t},is}} (rp^{\frac{k-1}{k}} - 1)}{\left( \frac{T\_3}{T\_1} \right) - 1 - \frac{1}{\mathfrak{m}\_{\mathfrak{t},is}} (rp^{\frac{k-1}{k}} - 1)}$$

The efficiency of the thermodynamic cycle depends mainly on the TIT (T3) or its dimensionless parameter θ= T3/T1 as well as the pressure ratio rp as shown in Figures 6b–d. The highest cycle temperature is limited by the material and cooling of the first turbine stages; pressure ratio can be optimized to maximize the efficiency for a specific combustor temperature. Besides optimization of the efficiency, the gas turbine is also optimized for power output (Figure 6d). The optimization sets the conditions for the combustor. For the gas turbine cycle in Figure 6 at a combustor outlet temperature at 1,743 K, the optimal pressure ratio for specific power is 14:3 bar and the optimal pressure ratio for efficiency is 25:1 bar. These values are engine specific but show the tendency for optimization. The efficiency of the thermodynamic cycle depends mainly on the TIT (T3) or its dimensionless parameter θ= T3/T1 as well as the pressure ratio rp.

The thermal efficiency always increases with the increase of θ or the TIT, T3, which has limitation with the materials. The pressure ratio rp (P2/P1=P3/P4) affects the cycle efficiency, which increases with rp until it reaches a maximum and then starts to fall. The optimal compression ratio changes with alteration of the compressor and turbine efficiencies.

The specific work, defined by the work per unit mass of the air, increases T3 and reaches a maximum for a certain rp as shown in Figure 6c.

Two distinct losses occur in the combustion chamber: combustion inefficiency and pressure loss.

The first implies an imperfect conversion of the chemical energy in the fuel/air mixture into thermal energy. It is defined as

$$
\eta\_{cc} = \frac{\overline{c}\_p \left[ (\dot{m}\_{air} + \dot{m}\_f) T\_{t3} - \dot{m}\_{air} T\_{t2} \right]}{\dot{m}\_f \Delta h\_u}
$$

The typical combustion efficiency is around 0.99 or better.

The thermal efficiency of a real gas turbine cycle is lower than the one of the ideal cycle. In the T-s diagram or in the P-v diagram, respectively, the differences are obvious since there are no more isentropic changes possible.

#### **2.3. Gas Turbine (GT) components**

GTs are operating according to Brayton cycle and using the following components.

#### *2.3.1. Air intake*

The air to compressor should pass through an air filter to prevent dust from entering the machine and is accelerated in a duct to the compressor. The inlet duct in front of the compressor is usually designed as a diffuser. This decelerates the air at the inlet and converts part of the air kinetic energy into pressure.

Figure 7a shows an air filter installed at the air inlet to the compressor. The inlet air duct can contain an air cooling system. The compression in the GT is a constant volume process. So, the air temperature decrease would increase the air density and mass flow rate, decrease the specific power consumed by the compressor (per unit mass), and increase the GT power output. Figure 7b shows the effect of compressor inlet temperature on the GT output power and heat rate. The air inlet temperature can be decreased by evaporative cooling, fogging, and chilled water system as shown on the psychometric chart given in Figure 7c.

Figure 8a shows an inlet air to compressor using evaporative cooling which used relative humidity and wet bulb temperature that are rather low. This system has the advantage of low capital and operation cost as it can operate on raw water and uses air washer that cleans the inlet air. Figures 8b–d show an inlet air to compressor using fogging system. It is also an evaporative cooling system that is used when relative humidity and wet bulb temperature are rather low. This system uses demineralized water and increases GT performance better than the previous evaporative cooling system.

Figure 9a shows mechanical refrigeration system (direct type) used hot in areas and can bring the air temperature to any specific requirement irrespective of ambient temperature and humidity ratio. This system has the advantage of increasing the GT performance better than evaporative cooling and fog system. However, this system has high initial capital cost and high operation and capital cost. Figure 9b shows the absorption refrigeration system (direct type), which is similar to that of Figure 9a, but with absorption cooling system operated mainly with steam or hot water substituting the mechanical refrigeration system. This systems has also the advantage of increasing the GT performance better than evaporative cooling and fog system, but at higher initial capital cost and high operation and capital cost.

#### *2.3.2. GT compressor*

The main parameters of a compressor are the required pressure ratio (rp), volumetric flow rate, consumed power, and permissible shaft length. The used compressors types in GT application are axial, centrifugal, and combination of both. Axial compressors have more stages to reach the same compression ratio achieved by centrifugal type, and thus, axial compressors have a longer shaft than centrifugal ones. Axial compressors have lower changes of flow direction during compression and thus better efficiency (82–90 %) compared to centrifugal (72–82 %). Axial compressors handle much wider range of volume flows, are used Cogeneration Power-Desalting Plants Using Gas Turbine Combined Cycle http://dx.doi.org/10.5772/60209 139

**2.3. Gas Turbine (GT) components**

air kinetic energy into pressure.

the previous evaporative cooling system.

*2.3.2. GT compressor*

*2.3.1. Air intake*

138 Desalination Updates

GTs are operating according to Brayton cycle and using the following components.

The air to compressor should pass through an air filter to prevent dust from entering the machine and is accelerated in a duct to the compressor. The inlet duct in front of the compressor is usually designed as a diffuser. This decelerates the air at the inlet and converts part of the

Figure 7a shows an air filter installed at the air inlet to the compressor. The inlet air duct can contain an air cooling system. The compression in the GT is a constant volume process. So, the air temperature decrease would increase the air density and mass flow rate, decrease the specific power consumed by the compressor (per unit mass), and increase the GT power output. Figure 7b shows the effect of compressor inlet temperature on the GT output power and heat rate. The air inlet temperature can be decreased by evaporative cooling, fogging, and

Figure 8a shows an inlet air to compressor using evaporative cooling which used relative humidity and wet bulb temperature that are rather low. This system has the advantage of low capital and operation cost as it can operate on raw water and uses air washer that cleans the inlet air. Figures 8b–d show an inlet air to compressor using fogging system. It is also an evaporative cooling system that is used when relative humidity and wet bulb temperature are rather low. This system uses demineralized water and increases GT performance better than

Figure 9a shows mechanical refrigeration system (direct type) used hot in areas and can bring the air temperature to any specific requirement irrespective of ambient temperature and humidity ratio. This system has the advantage of increasing the GT performance better than evaporative cooling and fog system. However, this system has high initial capital cost and high operation and capital cost. Figure 9b shows the absorption refrigeration system (direct type), which is similar to that of Figure 9a, but with absorption cooling system operated mainly with steam or hot water substituting the mechanical refrigeration system. This systems has also the advantage of increasing the GT performance better than evaporative cooling and fog system,

The main parameters of a compressor are the required pressure ratio (rp), volumetric flow rate, consumed power, and permissible shaft length. The used compressors types in GT application are axial, centrifugal, and combination of both. Axial compressors have more stages to reach the same compression ratio achieved by centrifugal type, and thus, axial compressors have a longer shaft than centrifugal ones. Axial compressors have lower changes of flow direction during compression and thus better efficiency (82–90 %) compared to centrifugal (72–82 %). Axial compressors handle much wider range of volume flows, are used

chilled water system as shown on the psychometric chart given in Figure 7c.

but at higher initial capital cost and high operation and capital cost.

**Figure 7.** a: Turbine inlet air cooling filter-house modification to place the cooling coil coming from ammonia compres‐ sion chiller plant [9] b: Typical inlet air cooling impacts on combustion turbine performance [1] c: Psychometric chart showing evaporative cooling process and chilled water cooling process [10]

in all heavy utility gas turbines, have much lower tendency for flow separation at the inlet blades, and are more reliable in the case of fast load changes. Centrifugal compressors have small-size, short shafts, used only in small gas turbines (less than 5 MW) and high rotor speeds. Combination of axial and centrifugal compressors utilizes axial compressor reliability and the centrifugal compressor high-pressure ratio.

In centrifugal compressor (Figure 10a) the air (to be compressed) enters the impeller center and moves outward by centrifugal force to the compressor discharge diffuser. The rotating impellers accelerate the air velocity, and the air kinetic energy is converted to an increase in static pressure by slowing the flow through a diffuser before being discharged.

**Figure 8.** a: An inlet air to compressor using evaporative cooling which used relative humidity and wet bulb tempera‐ ture that are rather low [11] b: An inlet air to compressor using fogging system which used relative humidity and wet bulb temperature that are rather low and using demineralized water [11] c: Fog system produces billions of microfine (10-micron average) droplets at 2,000 psi that create a much larger overall evaporative surface, which allows the drop‐ lets to evaporate and cool the airflow far more quickly than larger, heavier droplets. This results in faster, more effec‐ tive evaporation and cooling with significantly lower drain water rates [12] d: MeeFog™ array for a frame 7FA gas turbine, Mee Industries – Fogging Systems for Offshore Gas Turbines

(c) (d) **Figure 9.** a: Mechanical refrigeration system (direct type) used in areas where relative humidity is rather high [1] b: Absorption refrigeration system (direct type) used in areas where relative humidity is rather high [1]

(b)

(d)

Axial compressors have moving (rotor) and fixed (stator) blades (Figure 8b). The arrays of blades are set in rows, usually as pairs: one rotating and one stationary. While rotating airfoils (known as blades or rotors) accelerate the fluid, the stationary airfoils (known as stators or vanes) decelerate the air, i.e., slow it down, and its kinetic energy is converted to pressure energy. The stators redirect the flow direction for the rotor blades of the next stage. The discharge velocity is almost equal to the suction velocity. This process is repeated by several stages depending on the desired output pressure.

(a)

140 Desalination Updates

(b

)

(d

(b)

)

(c)

turbine, Mee Industries – Fogging Systems for Offshore Gas Turbines

(a)

(c)

**Figure 8.** a: An inlet air to compressor using evaporative cooling which used relative humidity and wet bulb tempera‐ ture that are rather low [11] b: An inlet air to compressor using fogging system which used relative humidity and wet bulb temperature that are rather low and using demineralized water [11] c: Fog system produces billions of microfine (10-micron average) droplets at 2,000 psi that create a much larger overall evaporative surface, which allows the drop‐ lets to evaporate and cool the airflow far more quickly than larger, heavier droplets. This results in faster, more effec‐ tive evaporation and cooling with significantly lower drain water rates [12] d: MeeFog™ array for a frame 7FA gas

(d) **Figure 9.** a: Mechanical refrigeration system (direct type) used in areas where relative humidity is rather high [1] b:

Absorption refrigeration system (direct type) used in areas where relative humidity is rather high [1]

**Figure 10.** a: Centrifugal-compressor flow, pressure, and velocity changes; (a) airflow through a typical centrifugal compressor and (b) pressure and velocity changes through a centrifugal compressor [13] b: Schematic diagram of an axial flow compressor and pressure and velocity profile [14]

The direction of flow is parallel to the direction of the rotation. The design of compressor blades is different than those of turbines. The compressor blades have divergent profile and act as diffuser to increase air pressure. The turbine blades have convergent profile which works as a nozzle, reducing air pressure by changing its pressure energy into kinetic energy. More on axial compressor design is given in Ref. [15]. Although an axial stage may not offer as much of pressure ratio as a centrifugal stage of the same diameter, a multistage axial compressor offers far higher pressure ratio (and therefore mass flow rates and resultant power) than a centrifugal design.

Separation of the air flow from the surface of the blades of the first compressor stage is real problem in axial and centrifugal compressors. Flow separation from the surface of single blades generates high turbulence in the grid and can partly block the flow path of the incoming air aerodynamically. This effect, called a rotating stall, stresses the whole gas turbine structure with oscillating pressure waves.

#### *2.3.3. GT combustor*

The compressed air leaving the compressor is directed to the combustion chamber (cc), called combustor, where fuel such as natural gas (or petroleum liquids) is injected. In a combustor (Figure 11a) the fuel chemical energy is converted to thermal energy. So, the combustor combines and mixes air and fuel, ignites them, and contains the mixture during combustion. The combustor contains basically four zones – primary zone, secondary zone, dilution zone, and various wall jets – to manage heat transfer at the combustor boundary as shown in Figure 11b. Air entering the combustor is distributed to four major injection points. The first is through swirl vanes positioned at the combustor front face and typically surround the fuel injection port. The swirl vanes impact a circumferential velocity component to the air and thereby thrust the air radially outward as the air enters the combustor (Figure 11c). This creates a pressure void at the center line and induces a backflow to fill the centerline pressure deficit. This effectively creates, as a result, a recirculation flow that extends approximately one duct diameter downstream and defines the "primary zone" of the combustor.

**Figure 11.** a: Stationary gas turbine electric power generator [16] Figure 11b: Schematic illustration of a general com‐ bustor [8] c: Circulation created by air swirler

The combustors are classified as:

The direction of flow is parallel to the direction of the rotation. The design of compressor blades is different than those of turbines. The compressor blades have divergent profile and act as diffuser to increase air pressure. The turbine blades have convergent profile which works as a nozzle, reducing air pressure by changing its pressure energy into kinetic energy. More on axial compressor design is given in Ref. [15]. Although an axial stage may not offer as much of pressure ratio as a centrifugal stage of the same diameter, a multistage axial compressor offers far higher pressure ratio (and therefore mass flow rates and resultant power) than a

Separation of the air flow from the surface of the blades of the first compressor stage is real problem in axial and centrifugal compressors. Flow separation from the surface of single blades generates high turbulence in the grid and can partly block the flow path of the incoming air aerodynamically. This effect, called a rotating stall, stresses the whole gas turbine structure

The compressed air leaving the compressor is directed to the combustion chamber (cc), called combustor, where fuel such as natural gas (or petroleum liquids) is injected. In a combustor (Figure 11a) the fuel chemical energy is converted to thermal energy. So, the combustor combines and mixes air and fuel, ignites them, and contains the mixture during combustion. The combustor contains basically four zones – primary zone, secondary zone, dilution zone, and various wall jets – to manage heat transfer at the combustor boundary as shown in Figure 11b. Air entering the combustor is distributed to four major injection points. The first is through swirl vanes positioned at the combustor front face and typically surround the fuel injection port. The swirl vanes impact a circumferential velocity component to the air and thereby thrust the air radially outward as the air enters the combustor (Figure 11c). This creates a pressure void at the center line and induces a backflow to fill the centerline pressure deficit. This effectively creates, as a result, a recirculation flow that extends approximately one duct

(c)

**Figure 11.** a: Stationary gas turbine electric power generator [16] Figure 11b: Schematic illustration of a general com‐

(b)

diameter downstream and defines the "primary zone" of the combustor.

(a)

bustor [8] c: Circulation created by air swirler

centrifugal design.

142 Desalination Updates

*2.3.3. GT combustor*

with oscillating pressure waves.


**Figure 12.** a: Annular combustion chambers [17] b: Gas turbine combustor arrangement [5] c: Several combustors ar‐ ranged equidistant on the same pitched circle diameter, and each consists of an inner flame tube or liner cylinder mounted on the same axis inside an outer casing cylinder, called tubular combustors [18]

The combustion process in the GT combustor can be classified as diffusion flame combustion or lean-premix staged combustion. In the diffusion flame combustion, the fuel/air mixing and combustion take place simultaneously in the primary combustion zone, and this generates regions of near-stoichiometric fuel/air mixtures where temperatures and NOx generation are very high. In lean-premix combustion, fuel and air are thoroughly mixed in an initial stage resulting in a uniform, lean, unburned fuel/air mixture which is delivered to a secondary stage where the combustion reaction occurs [19]. The combustion process starts with mixing the fuel with air supported by natural or forced turbulences in the airflow through the combustor. Continuous and stabilized combustion process is affected by the speed of fuel and air particles to the reaction zone, transport of flue gas from there, the speed of the chemical reaction in the reaction zone, and the residence time of any particle in the reaction zone. When the air-fuel mixing is slow compared to the chemical reaction rates, the mixing time controls the burning rate.

In diffusion flames, fuel and oxygen are mixed in the reaction zone through molecular and turbulent diffusion and have wide stability rate of combustion process. It has the advantages of relatively simple design of the fuel nozzles. Since the local conditions at the flame front are rich in fuel, diffusion combustion is insensitive against combustion instabilities and keeps on burning and generates regions of near-stoichiometric fuel-air mixtures with very high temperatures even at very lean conditions. The high temperature by diffusion flames leads to the production of large quantities of thermal NOx.

To reduce the reaction temperatures and/or the formation of thermal NOx, premix combustion is developed, where fuel and air are homogeneously mixed in an initial stage to become lean, unburned fuel-air mixture which is delivered to a secondary stage where the combustion reaction takes place. Manufacturers use different types of fuel-air staging, including fuel staging, air staging, or both; however, the same staged, lean-premix principle is applied. Gas turbines using staged combustion are also referred to as Dry Low NOX combustors. The majority of GT currently manufactured are lean-premix staged combustion turbines.

(c) **Figure 13.** a: Schematic of a conventional and a lean-premix combustor [20] b: Primary zone temperature influence on NOX and CO emissions [8]

In premix, mixing of fuel and air occurs far before the reaction zone. Depending on the burner design and the flow velocity, the time from the fuel injection to the moment of ignition is within several milliseconds. This time is used to create a mostly homogeneous mixture, with a fuel concentration within the ignition range of the specific fuel for the given compressor discharge temperature. The typical adiabatic flame temperature, to which a premix combustion system is adjusted, is at 1,750 K. At this temperature, the formation of NOx is still on an acceptable level, while the heat transfer from the flame is high enough to ensure the ignition of the fresh mixture (Figures 13a, b).

The combustion process in the GT combustor can be classified as diffusion flame combustion or lean-premix staged combustion. In the diffusion flame combustion, the fuel/air mixing and combustion take place simultaneously in the primary combustion zone, and this generates regions of near-stoichiometric fuel/air mixtures where temperatures and NOx generation are very high. In lean-premix combustion, fuel and air are thoroughly mixed in an initial stage resulting in a uniform, lean, unburned fuel/air mixture which is delivered to a secondary stage where the combustion reaction occurs [19]. The combustion process starts with mixing the fuel with air supported by natural or forced turbulences in the airflow through the combustor. Continuous and stabilized combustion process is affected by the speed of fuel and air particles to the reaction zone, transport of flue gas from there, the speed of the chemical reaction in the reaction zone, and the residence time of any particle in the reaction zone. When the air-fuel mixing is slow compared to the chemical reaction rates, the mixing time controls the burning

In diffusion flames, fuel and oxygen are mixed in the reaction zone through molecular and turbulent diffusion and have wide stability rate of combustion process. It has the advantages of relatively simple design of the fuel nozzles. Since the local conditions at the flame front are rich in fuel, diffusion combustion is insensitive against combustion instabilities and keeps on burning and generates regions of near-stoichiometric fuel-air mixtures with very high temperatures even at very lean conditions. The high temperature by diffusion flames leads to

To reduce the reaction temperatures and/or the formation of thermal NOx, premix combustion is developed, where fuel and air are homogeneously mixed in an initial stage to become lean, unburned fuel-air mixture which is delivered to a secondary stage where the combustion reaction takes place. Manufacturers use different types of fuel-air staging, including fuel staging, air staging, or both; however, the same staged, lean-premix principle is applied. Gas turbines using staged combustion are also referred to as Dry Low NOX combustors. The

(c)

**Figure 13.** a: Schematic of a conventional and a lean-premix combustor [20] b: Primary zone temperature influence on

(b)

majority of GT currently manufactured are lean-premix staged combustion turbines.

the production of large quantities of thermal NOx.

(a)

NOX and CO emissions [8]

rate.

144 Desalination Updates

In general, there is an operation window for low emissions that range from the primary zone temperatures 1,670 K to 1,900 K (Figure 13b). The upper temperature limit is set by the temperature dependence of NOX and the lower limit by carbon monoxide. The increase in CO for lower temperatures is related to poor combustion and the lean blowout limit for the burner.

**Figure 14.** a: A triple-stage turbine with single-shaft system [17] b: The gas turbine section of the Siemens V94.2 gas turbine. c: Turbine stage with stators to the left which have the main function to act as nozzles to increase the velocity of the gas primarily in the tangential direction, by converting pressure energy to kinetic energy. To the right of the stators are the rotors, which have the function to convert the kinetic energy to power by causing a rotation of the shaft [4] d: Temperature and pressure throughout gas turbine [18]

### *2.3.4. GT turbine*

The hot gases produced in the combustor are expanded in the turbine (Figures 14a–d) to give mechanical energy that operates the compressor, and the balance produces the electric power (EP). The turbine, similar to the compressor, can be axial or centrifugal type. The axial type is easier to cool, as the turbine is exposed to high thermal stresses by the hot gases entering the turbine. The turbine cooling is crucial as it provides the potential of raising the TIT and thus the efficiency. Gas turbines can be particularly efficient when heat content of the hot gases from the turbine is recovered in HRSG to power a conventional ST in GTCC. The hot gases from the GT can also be used for space or water heating or drive an absorption chiller for cooling the inlet air and increase the power output. Figure 14d shows that the hot gases leaving the GT are high enough to generate steam.
