Applied Energy 80 (2005) 261272

APPLIED ENERGY
www.elsevier.com/locate/apenergy

Inuence of ambient temperature on combined-cycle power-plant performance

Felipe R. Ponce Arrieta *, Electo E. Silva Lora
Centro de Excel^ncia em Gera ^o Termeltrica e DistribuidaNEST, Universidade Federal de Itajub, e ca e a Av BPS 1303, CP 50, Itajub, MG, CEP 37 500-903, Brazil a Accepted 11 April 2004 Available online 7 June 2004

Abstract Thermal-power electricity are important mainly because of the need for diversied powergeneration and the availability of natural gas, the main fuel used in this type of electricitygenerating system. With the implementation of the priority plan for thermal-power plants in Brazil, dozens of units will be installed which will make it possible to transform the Brazilian electric system, today mainly based in hydraulic principles, into a hydro-thermal system. The operation of a combined cycle thermal-power plant is inuenced by the conditions that are present at the place where it is installed, mainly ambient temperature, atmospheric pressure and the airs relative-humidity. These parameters aect the generated electric-power and the heat-rate during operation. Among these variables, the ambient temperature causes the greatest performance variation during operation. That is the reason why the inuence of this variable on this type of generating unit is studied. The plant selected for this study has a multiple-shaft conguration and is composed of two Siemens AG 501F gas-turbines, coupled to three pressure levels HRSGs and re-heating with supplementary ring and a steam-turbine. The most relevant results obtained from a thermodynamic simulation, in which the Gate Cycle Software version 5.51.0.r was used, are the curves of generated power, as well as the heat rate and thermal eciency as functions of ambient temperature and the supplementary ring. 2004 Elsevier Ltd. All rights reserved.
Keywords: Electric-power generation; Operation; Thermodynamic analysis; Eciency; Heat rate

Corresponding author. Tel./fax: +55-35-3629-1355. E-mail address: fponce@unifei.edu.br (F.R.P. Arrieta).

0306-2619/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.apenergy.2004.04.007

262

F.R.P. Arrieta, E.E.S. Lora / Applied Energy 80 (2005) 261272

Nomenclature BOP GT HP HPST HR HRSG IP IPST ISO LHV LP LPST NG ST balance of plant gas turbine high pressure high pressure steam turbine heat rate heat recovery steam generator intermediate pressure intermediate pressure steam turbine International Organization for Standardization lower heating value low pressure low pressure steam turbine natural gas steam turbine

1. Introduction Thermal-power plants began to gain strength in Brazil after the demand for diversication of the generating capabilities and the availability of natural gas, which is the main kind of fuel used by this type of generating unit. The implementation of the priority thermal-power electricity program will result in the installation of several thermal-power plants. This will make the Brazilian electric system become hydrothermal rather than predominantly hydroelectric. The operation of thermal-power plants is considerably more complex than the operation of hydroelectric plants, because of the use of working uids at high temperature and pressure and the consequent dicult operational conditions of the metal tubes, heating surfaces, turbines combustion-chamber, casing, headings, etc. In addition, one must also consider the inuence of metal corrosion, the need for complex automatic control systems, the need to implement pollution control, and the eort needed for maintaining operational high eciency and reliability. Currently, the use of natural gas for power generation considering its cost is justied, in most cases, by high-eciency installations, which are typical of combined-cycle thermal-power plants. In this type of generating unit, the operation problems pointed out above manifest themselves, and they are aggravated as a consequence of the presence and the interconnections of the main components, namely the gas turbine, HRSG and steam turbine. Besides these operational problems, those units, designed for ISO conditions (15 C ambient temperature, 101.32 kPa atmospheric pressure and 0.6 air relative-humidity) are extremely sensitive to changes in ambient conditions. The main reason for this sensitivity is the inuence of these parameters on o-design gas-turbine operation (namely ambient

F.R.P. Arrieta, E.E.S. Lora / Applied Energy 80 (2005) 261272

263

temperature, atmospheric pressure and relative humidity are dierent from those under ISO conditions), where, on average, about two-thirds of the plants total power is generated. Nevertheless, it is possible to say that combined-cycle thermalplants with similar congurations present similar o-design behavioral trends, although dierent capacities have been designed [1]. While the air temperature increases in relation to the design temperature, the power generated by the plant decreases. This is explained by the considerable inuence that this parameter has on the power generated by the gas turbine. According to Kehlhofer et al. [1], the gas turbine is designed to operate with a constant air volume in the compressor. When the ambient-air temperature increases, its specic mass is reduced. In order to ensure the same air volumetric ow, the mass ow is reduced, so causing the power of the gas turbine and the amount of heat generated in the HRSG to fall. The inuence of the atmospheric pressure is related to the air-density variation. For low pressures, that is, high altitudes in relation to the sea level, the air density decreases. Disregarding the pressure losses in the gas-turbines inlet and outlet ducts, and considering that the eciency of the steam cycle does not change, the combined-cycle plant presents a behavior that is similar to what was previously explained for ambient temperature. The eect of the ambient pressure on the performance must be considered, mainly during the design phase, for once the plant is installed, the variations of this variable are neglected. While the airs relative-humidity increases, the power generated by the combined-cycle plants also increases, provided the other parameters remain constant. In this case, the gas-turbines eciency is slightly reduced, as well as its power. However, the temperature of the gas-turbines exhaust-gases rises, and therefore the power generated by the steam cycle is increased. The nal result depends on which of the factors described above is predominant, but in both cases, the total-power variation is very small. Plants with cooling towers deserve special attention. In these plants, the airs relative-humidity is directly related to the vacuum in the condenser, and consequently to the temperature of the steamturbines exhauststeam. In these cases, a lower air relative-humidity results in a greater vacuum and a higher eciency. The analysis previously presented allows the supposition that out of these three variables, the ambient temperature has the greatest inuence on the o- design operation of a combined-cycle plant. Therefore, the main goal of this study is to evaluate the inuence of this variable on the operation of this type of installation. In order to carry out this study, the values of the other variables were kept constant, for example, atmospheric pressure, airs relative-humidity, electric frequency, power factor and fuel characteristics.

2. Assumptions The simplied thermal scheme of the installation selected for the analysis is shown in Fig. 1. The thermal schemes main characteristics are: Two Siemens AG 501F gas-turbines.

264

F.R.P. Arrieta, E.E.S. Lora / Applied Energy 80 (2005) 261272

Fig. 1. Combined-cycle plant simplied thermal-scheme.

Two HRSGs presenting three pressure-levels and fuel supplementary ring. A detailed scheme of the HRSG is displayed in Fig. 2. According to this gure, fuel supplementary-ring takes place after the two nal high-pressure superheating stages. The rst re-heating stage and the rst high-pressure superheating stage are placed after the supplementary-ring stage. A high, intermediate and low-pressure steam-turbine: the last one has a double ow down exhaust. Dearator condenser with a cooling system that has a wet-tower. Feed-water pumps: the low-pressure pump at the condenser outlet and the highpressure pump are responsible for elevating the water pressure to high and intermediate levels. Natural-gas supply: the fuel that will be used by the gas turbines is heated, but the fraction that will be used for the supplementary ring is not.

Fig. 2. HRSG detailed scheme.

F.R.P. Arrieta, E.E.S. Lora / Applied Energy 80 (2005) 261272

265

Fig. 3 illustrates the thermal scheme of a combined-cycle drawn by using Gate Cycle version 5.51.0.r: this gure also presents the results of the design-point simulation. The most important parameters of the installation described above are presented in Table 1 and refer to parameters used for the thermodynamic simulation at the design point. It is possible to observe that the data shown in Table 1 are divided into four main groups. In the rst group, besides the ISO parameters, the net total electric-power generated by the plant at the design point is 600 MW. The second group of data denes the fuel used during the simulation, as well as its low caloric-value and the supply conditions. The third group is dedicated to the gas turbine. In this case, the data presented had already been implemented in the Gate Cycle software, when the turbine was selected. The last group is related to the steam cycle and the plants auxiliary systems. The following values can be observed: steam pressure and temperature for dierent levels of pressure presented by the HRSG, condenser operating pressure, eciency of the steamturbines three stages and the imposition of minimum steam quality at the lowpressure turbines outlet, which is based on technical criteria. The other data that were shown are related to the fuel supplementary-ring, the condensers cooling system, the pumps and the BOP power consumption. It is important to highlight that the temperature of the gases after the fuel supplementary ring was limited to 675 C, so avoiding the formation of steam in the nal section of the economizer tubes. The nal suppositions are related to the calculation of properties during the simulation. In this sense, water and steam properties were determined according to Reynolds [2], whereas an ideal-gas behavior was considered for the gases, and the proprieties were calculated according to Chase [3].

02 GT - 501F Siemens AG + 01 Steam turbine

Fig. 3. Thermal scheme of a combined-cycle plant.

266

F.R.P. Arrieta, E.E.S. Lora / Applied Energy 80 (2005) 261272

Table 1 Main data for the simulation of the combined-cycle thermal plant at the design point Parameter (unit) Ambient temperature (C) Atmospheric pressure (kPa) Relative humidity Net total electric-power (MW) Fuel Natural gas, LHV (kJ kg1 ) Supply conditions (MPa/C) Siemens AG 501F gas-turbinesa Gross power (MW) Maximum compression isentropic-eciency Combustion eciency Turbines isentropic-eciency Electricity-generators eciency Turbines inlet-temperature (C) Turbines outlet-temperature (C) Cooling air fraction Auxiliary power consumption (MW) Steam cycle with three pressure levels and re-heating High-pressure steam (MPa/C) Intermediate-pressure steam and re-heating (MPa/C) Low-pressure steam (MPa/C) Condensers operating-pressure (kPa) HRSGb Gas-temperature after supplementary ring (C) Supplementary ring Minimal gas-exiting temperature (C) Heat-transfer coecient (kJ s1 m2 K1 ) Three-stage steam turbine Net power with additional burning (MW) High-pressure turbines isentropic-eciency Intermediate-pressure turbines isentropic-eciency Low-pressure turbines isentropic-eciency Minimum quality at the outlet of the low-pressure steam turbinec Total electromechanical eciency Condenser Heat-exchange area (m2 ) Heat-transfer global coecient (kJ s1 m2 K1 ) Cooling tower Capacity (kJ s1 ) Number of fans Fan total power (kW) BOP Pump isentropic eciency Total losses referring to the steam turbine power Value 15 101.32 0.60 600 46515 2.758/15 174.66 0.90 0.99 0.9431 0.985 1382.5 608 0.178 1.18316 15.6/530 3.2/530 0.75/305 5.06 675 0.976 70 0.45426 253.57 0.8098 0.9259 0.8867 0.85 0.94464 30,528 2.85 421,418 10 960.5 0.75 0.0198

F.R.P. Arrieta, E.E.S. Lora / Applied Energy 80 (2005) 261272

267

Table 1 (continued ) a Source: [4]. b The temperature of the gas after the supplementary ring was limited to 675 C, so aiming at avoiding the formation of steam in the nal section of the economizer tubes during the o-design operation. The pressure drop of the gas and of the working uid (watersteam) was not considered. c According to Boyce [5], this value avoids blade erosion within the last stages of the turbine.

3. Methodology This section explains the criteria used to evaluate the inuence of the ambient temperature on the operation and performance of a combined-cycle plant and the dierent stages in which the study was carried out. For the particular case of the installation presenting the characteristics shown previously, it is not possible to disregard the eect of the fuel supplementary-ring on the plants performance. By using the fuel supplementary-ring in the HRSG, it is possible to mitigate the power loss in the gas-turbines caused by a rise in the ambient temperature, so aecting the installations thermal-eciency. The power loss in the gas-turbines is compensated by the supplementary-ring, for it is possible to generate more steam in the HRSG, and consequently, more power in the steam turbine. Thus, aiming at accomplishing this studys goal, a parametric study involving two variables (ambient temperature and gas temperature after supplementary ring) was carried out. Table 2 shows the values of the variables that were used for the parametric study, that is, for each value of ambient temperature that was considered, the plants performance was evaluated using dierent gas-temperatures after the supplementary ring in the HRSG. The parametric study was carried out according to the following stages: Stage 1. Selection of the installations thermal scheme. In this case the choice was a typical scheme of a combined-cycle plant for the generation of a net electricpower of 600 MW. Stage 2. The thermal scheme was drawn using the Gate Cycle software. It includes the components geometries, distribution and their interconnection. Stage 3. Data input of all the components that form the thermal scheme: gas turbines, HRSGs surfaces, steam turbine, condenser, equipment and auxiliary subsystems.

Table 2 Parametric study of the variables and their values Variable Ambient temperature (C) Gas temperature after the supplementary ring (C) Considered values 0, 5, 10, 15, 20, 25, 30 and 35 675, 645, 615, 585, 555 and 525

268

F.R.P. Arrieta, E.E.S. Lora / Applied Energy 80 (2005) 261272

Stage 4. The scheme and the input data must be tted to ISO conditions. At this stage, the program was run several times and the thermodynamics, physics, geometric data and mathematic parameter thermodynamics errors were analyzed. The main goal of this stage is to make the scheme ready for the o-design simulation. Stage 5. Simulation of the o-design operation for each of the dierent specied ambient temperature values, maintaining the gas temperature after the supplementary ring within project conditions (i.e., <675 C). Stage 6. Simulation of the o-design operation for each of the dierent ambient temperature values, and varying the gas temperature after the supplementary ring according to the specied values. As a result of stages 5 and 6, 48 operation variants were achieved. Stage 7. Verication of results. The analysis of the limit values of dierent variables was carried out. This analysis aims at avoiding reaching risky values that could jeopardize the operations reliability, for example, steam at excessively high temperatures at the high-pressure turbine inlet. Stage 8. Presentation of graphics to analyze the results. Stage 9. Results analysis as discussed in Section 4.

4. Results and analysis In order to evaluate the performance of a combined-cycle plant during o-design operation, because of changes in the ambient temperature and the gas temperature after the supplementary ring, variations in the generated power, eciency and heat rate were analyzed. The combined cycle plants net thermal eciency gNet is given by gNet WGT1 kW WGT2 kW WST kW WAUXILIARY kW 100%; _ mFUEL kg=s LHVkJ=kg 1

where W represents the gross power-generated by the gas turbines GT1 and GT2, _ the steam turbine ST, and the BOP power consumption AUXILIARY: m represents the fuels mass-ow consumed in the plant and LHV its low heating-value. The combined cycle plants net heat-rate HRNet is given by HRNet 3600 100kJ=kW h; gNet % 2

where the net thermal eciency gNet is calculated according to Eq. (1). Fig. 4 shows the variation in the net power generated in the combined-cycle plant in relation to the gas temperature after the supplementary ring for the dierent ambient temperatures that were analyzed. According to this gure, the results of the simulation are: For any ambient temperature, the operational performance tendency is the same. The rise in the temperature of the gas after the supplementary ring increases the

F.R.P. Arrieta, E.E.S. Lora / Applied Energy 80 (2005) 261272

650

269

600 Net power, MW

550

500

450 525 555

Ta = 0C

585 615 645 Temperature after the supplementary firing, C

Ta = 5C

675

Ta = 10C

Ta = 15C

Ta = 20C

Ta = 25C

Ta = 30C

Ta = 35C

Fig. 4. Net electric-power generated.

power generated in the plant. This behavior is caused by the increase in the power generated in the steam turbine. For the studied gas temperature range after the supplementary ring, it is possible to generate up to 70 MW more power in the installation. The change in the studied parameters has a signicant inuence on the power that can be generated in the installation. For the maximum ambient-temperature and the minimum gas-temperature after the supplementary ring, the generated power is 468 MW, whereas for the minimum ambient-temperature and the maximal gastemperature after the supplementary ring, the generated power is 642 MW, that is, a variation of 170 MW in the generated power was obtained. Fig. 5 illustrates the variation in the net power-generated in the gas and steam cycle of the combined cycle plant in relation to the ambient temperature. The dark part of the graph shows the net power generated in the gas cycle, whereas the cross part shows the net power generated in the steam cycle. The evident result is that the sum of the two net powers generated in both cycles is equivalent to the net power generated by the combined-cycle plant. In this graph, the eight peaks that were observed correspond to the condition of maximum gas temperature after the supplementary ring for each of the values of the ambient temperature that were analyzed, whereas the valleys correspond to the condition of minimum gas-temperature after the supplementary ring. The following observations can be made: The strong inuence of ambient temperature produces a fall in the power generated in the gas cycle from 380 to 305 MW, that is, 75 MW approximately. By using fuel supplementary-ring, for the whole range of temperature after the ring that was analyzed, it is possible to have a power gain in the steam cycle of approximately 77 MW.

270

F.R.P. Arrieta, E.E.S. Lora / Applied Energy 80 (2005) 261272

Fig. 5. Net electric-power generated in the gas and steam cycles.

The use of supplementary ring in a combined cycle installation allows a signicant compensation for the falls in the power generated in the gas cycles caused by variations in the ambient temperature. Fig. 6 shows the region where the combined-cycle plants thermal-eciency can be found in relation to the variables considered for this study. According to this gure, the following were observed: The highest eciency values are registered when the ambient temperature is least. This is the result of an increase in the power generated by the gas cycle because of the ambient temperature reduction. For a temperature of 0 C, the eciency variation lies in a range between 53.5% and 55.5%;

Fig. 6. Net eciency.

F.R.P. Arrieta, E.E.S. Lora / Applied Energy 80 (2005) 261272

660 6950 6900 6850 610 6750 560 6700 6650 6600 510 6550 6500 460 0 5 10 15 20 25 Ambient temperature,C 30 35 6450 Heat Rate, kJ/kW-h 6800 Net power, MW

271

Maximum temperature after the supplementary firing Minimium temperature after the supplementary firing

Fig. 7. Power and heat rate.

The highest eciency values were registered when the gas temperature after the supplementary ring is small. This behavior takes place due to a reduction in fuel consumption in the HRSG burners in order to achieve a lower gas temperature. Supplementary ring causes, on average, a fall in eciency of 1.5% for any of the ambient temperature values that were studied. The combined-cycle thermal eciency variation lies within the range between 52% and 55.4%. Fig. 7 shows the results of the parametric study as a function of the analyzed variables, so highlighting the eect of supplementary ring. According to this gure, it is possible to make the following observations: While the ambient temperature rises, the net power generated in the combined-cycle thermal-plant decreases in spite of the use of the maximum supplementary-ring. It was registered that, with a gas temperature of 675 C after the supplementary ring, the net electric-power varies in the range from 640 to 540 MW when the ambient temperature varies between 0 and 35 C; While the ambient temperature rises, the combined cycle thermal plants heat-rate increases (that is, the eciency decreases), in spite of the use of the minimum supplementary-ring temperature. The heat rate is even greater when the maximum supplementary-ring temperature is used.

5. Conclusions The parametric studied carried out to evaluate the inuence of the ambient temperature on the operation and performance of combined-cycle plants allows the following conclusions to be drawn:

272

F.R.P. Arrieta, E.E.S. Lora / Applied Energy 80 (2005) 261272

The ambient temperature has a signicant inuence on this type of generating unit. Within the range of values that was analyzed, a variation in the net power of the gas cycle of approximately 75 MW was registered. The ambient temperature and the temperature of the gas after the supplementary ring produce dierent eects on the combined-cycle plants performance. A drop in the ambient temperature increases the electric-power generated in the plant and its eciency as well, and vice-versa. The rise in the temperature of the gas after the supplementary ring increases the generated electric-power but reduces the eciency, and vice-versa. The variation of these two variables led to a variation in the generated net electric-power of 170 MW, and a variation in eciency of 3.4%. The supplementary ring is a technological alternative that can mitigate the power reduction in the gas cycle caused by a rise in the ambient temperature. However, it is evident that this alternative utilization reduces the thermal eciency. This way, the project of combined cycle thermal-plants with supplementary ring in HRSGs is restricted to installations where expenses with investment and fuel are lower than the income attained by selling the surplus generated energy. In other words, the use of supplementary ring must be analyzed taking the economic context into account. In this way, it is possible to compare its positive eects (i.e., mitigation of power reduction due to a rise in ambient temperature and/or the generation of surplus power for the market) with its main negative eect, that is the fall in the cycles eciency. Finally, it is important to highlight that the operation of combined-cycle thermal plants, is very complex, and that this study is far from establishing denitive criteria.

Acknowledgements The authors would like to thank the National Council for Scientic and Technological Development-CNPq for the nancial support granted for this study.

References
[1] Kehlhofer RH, Warner J, Nielsen H, Bachmann R. Combined cycle gas-steam turbine power plants, Ed. Pennwell, USA; 1999. p. 288. [2] Reynolds WC. Thermodynamics properties in SI. Stanford University Department of Mechanical Engineering; 1979. [3] Chase Jr MW. NIST-JANAF Thermochemical tables. 4th ed. American Institute of Physics; 1998. [4] Gas-Turbine World Handbook. Faireld, CT, USA: Pequot Publishing, 1998. [5] Boyce M. Performance monitoring of large combined-cycle power plants. In: Joint Power Generation Conference; vol. 2, PWR vol. 34. ASME; 1999. pp. 183190.