Modeling of combined cycle power plant based on a genetic algorithm parameter identification method

Classic combined cycle power plant models are often too complex for power system dynamic analysis, and hard to estimate the parameters accurately. The performances of parameter identification procedures have been significantly reduced by high-dimensional searches and strong nonlinear relationships. A new non-linear model is proposed to be suitable for the parameter identification procedure in this paper. An identification method based on an improved genetic algorithm (GA) is used for the modeling of a 400MW combined cycle power plant. The whole system is divided into six parts and an artificial disturbance is recorded for the identification of each part. The results show great consistence between identified model responses and the experimental data.