基于RBF神经网络的高压断路器机构故障诊断

High-Voltage Circuit Breaker Fault Diagnosis Based on RBF Neural Network

为诊断高压断路器操作机构故障,分析高压断路器机构故障时的分合闸线圈电流,本文通过提取时间和电流特征参数,对故障特征参数进行相对归一化处理后输入RBF神经网络,建立基于果蝇—粒子群混合算法的高压断路器RBF神经网络模型,用于高压断路器操作机构故障识别。以MATLAB为实验平台,通过训练样本和测试样本的仿真分析,得出RBF神经网络的输出结果与期望输出一致,实验验证该方法能有效实现高压断路器机构故障诊断,且诊断速度快、准确率高,具有较为广阔的应用前景。

In order to diagnose the fault of the operating mechanism of the high voltage circuit breaker,the current of the on-off coil during the fault of the high voltage circuit breaker is analyzed,This paper extracts the time and current characteristic parameters,and the characteristic parameters of the fault are relatively normalized and then input into the RBF neural network,a RBF neural network model for High Voltage Circuit Breaker(HV circuit breaker)was established based on a Hybrid Algorithm of particle swarm optimization(PSO).Using Matlab as the experimental platform,through the simulation analysis of training samples and test samples,reach the results of RBF neural network and network are consistent with the expected output.The experiment proves that this method can effectively realize the fault diagnosis of high voltage circuit breaker mechanism,and the diagnosis speed is fast,the accuracy rate is high,has the broader application prospect.