Study on simulation model of switched reluctance startor/generator system based on wavelet neural network

This paper presents a novel integrated simulation model of switched reluctance startor/generator system based on wavelet neural network. The switched reluctance startor/generator system is hard to obtain good simulation model due to its critical nonlinearity. Relatively accurate mathematical model would be built by the wavelet neural network due to its powerful learning, approximation and prediction abilities. The nonlinear mapping relation of the flux-linkage, current and rotor position has been realized by the wavelet neural network and the simulation model of switched reluctance startor/generator system is established meanwhile. The simulation has high operation speed and can be modified conveniently. The effectiveness of the proposed simulation model has been confirmed both by simulation and experiment results.