基于概率神经网络的高压断路器故障诊断模型

针对高压断路器在线监测与故障诊断问题,提出了一种基于概率神经网络的高压断路器故障诊断模型.该模型利用Parzen窗函数和Bayes分类规则建立了前向型自监督神经网络,在分析分（合）闸线圈信号的基础上提出了高压断路器故障模型.仿真结果表明,该模型对断路器的故障模式识别过程具有训练速度快、输出误差小和收敛性好等特点.

High Voltage Circuit Breaker Fault Diagnosis Model Based on Probabilistic Neural Networks

The high-voltage circuit breaker is the most important control and protection equipment in electrical power system,so on-line monitoring and fault diagnosing are the one of the most important parts in keeping the power grid normal operation function.This paper presents a high-voltage circuit breaker fault diagnosis model based on the probabilistic neural network.This model builds a feed-forward self-supervisory neural network based on the Parzen window function and Bayes classifying rules,and sets up a high-voltage circuit breaker fault diagnosis model on the basis of analyzing time signal of breaking coil.The simulation experiment results indicate that this model has the ability of fast convergence,small output error,good discrete and other characteristics in the process of the fault pattern of the circuit breaker.