Optimal waste heat recovery in micro gas turbine cycles through liquid water injection

Abstract

Listen

Translate article

Water injection in the compressor exhaust, to recuperate waste heat, is considered a possible route to improve the electric efficiency and overall performance of the micro Gas Turbine turbine (mGT). Many research exists on water injection in mGTs, however a generic study to determine the optimal route for waste heat recovery is still missing. To determine the optimal cycle settings for waste heat recovery through water injection, we have performed simulations using a two-step method. In a first step, the thermodynamic limit for water injection is sought using a black box method. In a second step, the cycle layout is designed by means of composite curve theory.This paper summarizes the results of two scenarios. In the first scenario, the black box is considered as adiabatic and no fixed stack temperature is imposed (thus allowing condensation of the exhaust gasses). One of the major concerns when injecting water is the water consumption, which can be compensated in some cases through condensation and recycling the condensate. Therefore, in the second scenario, the cycle is made self-sufficient with water. In this case, the black box is no longer considered adiabatic and heat exchange with the environment is allowed for condensation of the flue gasses.Black box simulations showed that lowering the stack temperature to 53 °C results in an injection of 17 %wt of water and an increase in electric efficiency of 9% absolute. To keep the mGT cycle layout simple, low cost and not too complex, a maximum of two heat exchangers was imposed for the heat exchanger network design. Although black box analysis indicated a large potential for water introduction, this potential could not be achieved with the considered networks in this paper. Finally, injection of preheated water was identified as the optimal water injection scheme for waste heat recovery resulting in 4.6% absolute electric efficiency increase and a final stack temperature of 62 °C. Results of simulations of the second case indicate that the stack temperature needs to be lowered under 26 °C in order to make the cycle self-sufficient with water.