基于支持向量机的高压断路器故障诊断方法的研究

Research on fault diagnosis method of high voltage circuit breaker based on support vector machine

基于支持向量机的故障诊断技术逐渐成为当前故障诊断领域研究的热点问题,而高压断路器作为关系电力系统可靠性的重要设备,其故障与否直接关系到整个电力系统的稳定.本文通过比较多种核函数的多重参量,最终选择适合高压断路器的核函数;通过分析传统交叉验证法,提出改进型交叉验证(K-CV)法,最后利用高压断路器现场实际数据作为样本,采用Libsvm工具箱,通过不断地参数寻优,最终得出故障分类结果.实验结果表明,采用的基于径向基核函数改进型10-CV交叉验证法支持向量机能更快速、准确的判断高压断路器故障类型.

Fault diagnosis technology based on support vector machine has become a hot research issue in the field of fault diagnosis of high voltage circuit breaker, and as an important equipment of power system reliability, it is directly related to the stability of the entire power system. In this paper, with the comparison of multiple parameters of the linear kernel function, polynomial kernel function, radial basis kernel function and multi layer perceptron kernel function, the final choice is the function for high voltage circuit breaker. Through the analysis of the traditional cross validation method, this paper presents an improved cross validation（K-CV） method, the use of high voltage circuit breaker field actual data as the sample, using the through the Libsvm toolbox, and constantly search parameters, finally obtains the fault classification results. The experimental results show that the improved 10-CV cross validation method based on radial basis kernel function support vector machine prove to be more rapid and more accurate in judging the fault of high voltage circuit breaker type.