Modeling of a Combined Cycle Gas Turbine (CCGT) Using an Adaptive Neuro-Fuzzy System

The multiple advantages of the combined power cycles make their improvement a priority. The modeling of a Combined Cycle Gas Turbine (CCGT) for the identification/recognition of operating variables that can provide an increase in efficiency is a challenge in this type of thermal system. The simulation was carried out using an adaptive neuro-fuzzy inference system (ANFIS). The influence of three input variables on the thermal efficiency of the combined cycle was analyzed: the pressure ratio (compression ratio) used in the gas cycle, the pressure of the steam extraction for preheating of feed water (if applicable) and the heat lost to the outside in the steam turbines. It is shown that the ratio of pressures in the gas cycle has the most significant effect on the efficiency. The optimal value of pressure in steam extraction was obtained, corresponding to the maximum efficiency of the cycle, equal to 62.38%. A comparison of the efficiency values obtained using ANFIS and the results of parametric analysis showed their insignificant discrepancy. The proposed approach using ANFIS can be an alternative to the usual phenomenological model when modeling the operating modes of the combined cycle.