Thermal modeling of a solid oxide fuel cell and micro gas turbine hybrid power system based on modified LS-SVM

摘要: Abstract For a solid oxide fuel cell (SOFC) integrated into micro gas turbine (MGT) hybrid power system, SOFC operating temperature and inlet are the key parameters, which affect performance of system. Thus, least squares support vector machine (LS-SVM) identification model based on an improved particle swarm optimization (PSO) algorithm is proposed to describe nonlinear dynamic properties SOFC/MGT system in this paper. During process modeling, PSO employed optimize parameters LS-SVM. In order obtain training prediction data identify modified LS-SVM model, physical established via Simulink toolbox MATLAB6.5. Compared conventional BP neural network standard LS-SVM, simulation results show that can efficiently reflect response