Research on Analysis of Power System Transient Signal by Neural Network and Genetic Algorithm

It is an important and difficult subject to acquire accurate transient signal parameters in the field of power system signal analysis. A neural network combined with genetic algorithm analysis approach for power system transient signal is proposed to improve the analysis precision of the transient signal including a period component varying in spectral waveform, a non-period component varying in exponential waveform and switching operation sequence. The signal parameters which are assembly estimated based on numeric differential and genetic algorithm are used as initial value of the neural network. The fundamental frequency is treated as weight to be adjusted to estimate the signal frequency and all harmonics' amplitudes and phases. The learning rate and the delayed frequency adjustment improve the convergence performance. The result of Matlab simulation demonstrates that the algorithm achieved high accuracy and rapid convergence.