Prediction of synchronous closing time on permanent magnet actuator for vacuum circuit breaker based on PSO-BP

In order to improve the prediction accuracy of phase controlled switching operation time, restrain overvoltage and inrush current when the circuit breaker switched on, this paper establishs BP network prediction model based on voltage and ambient temperature as the main input parameters, and weights the uncertainty influence factors such as aging and wear. In order to improve the accuracy of prediction model, proposing a method of BP neural network based on particle swarm optimization (PSO), comparing network prediction performance before and after algorithm optimized. The research results show that use the BP neural network based on PSO is more accurate than the prediction results which only by BP neural network predicts, the error of BP neural network based on PSO can control the predictive error within 0.2% and meet the requirement of synchronization control.