A new approach to determine subtransient parameters of synchronous machine using wavelet transform and artificial neuron network

摘要

Based on the wavelet transform and artificial neuron network, a new method is presented to determine parameters of synchronous machine from sudden line-to-line short-circuit test. An appropriate wavelet function is chosen to analyze the sudden line-to-line short-circuit transient current and the DC and fundamental component of the transient short-circuit currents are obtained by the wavelet transform. Compared with the traditional methods, the analysis using wavelet transform is more effective. Thus, the precise transient parameters and aperiodic component time constant are estimated by the artificial neuron network. At last, the simulation of sudden line-to-line short-circuit are carried out in a three-phase synchronous machine. The simulation results show that the method developed in this paper is valid.