基于RBFNN的高压断路器机械故障诊断系统

高压断路器是电力系统中最重要的控制和保护设备,其故障检修和诊断一直是电力运行部门高度重视的问题。为了提高高压断路器故障诊断的效率和准确率,提出了高压断路器机械故障诊断的径向基函数神经网络(RBFNN)方法;并根据其基本原理建立了高压断路器操动机构故障诊断的RBFNN模型;利用Matlab工具,使用来自现场的实际数据,通过故障诊断仿真实例,分析、验证RBFNN模型的性能,并对不同方法进行了对比分析。结果显示RBFNN训练速度快、逼近误差小,对输入输出关系比较复杂的高压断路器操动机构的故障诊断有很高的判断效率和准确率。

Diagnosis system for mechanical faults of HVCB based on RBFNN

High Voltage Circuit Breaker (HVCB) is one of the most important equipments in power system for controlling and protecting. In order to increase the efficiency and veracity in the fault diagnosis for HVCB, a method for diagnosing mechanical faults of HVCB was introduced based on radial basis function neural networks (RBFNN), and a diagnosis model was built. Based on this, with Matlab tools and data from field, simulation for fault diagnosis was realized to verify the RBFNN model. The results show that the method has a rapid training speed, and very little approach error, which can diagnose the mechanical faults effectively, and can be used to analyze the complex mechanical fault of HVCB.