基于小波包能量熵与SVM的微电网故障诊断  

Microgrid Fault Diagnosis Based on Wavelet Packet Energy Entropy and SVM

作　　者：戴连铭 李春华[1] DAI Lianming;LI Chunhua(Jiangsu University of Science and Technology,Zhenjiang 212000)

机构地区：[1]江苏科技大学,镇江212000

出　　处：《计算机与数字工程》2021年第10期2126-2132,共7页Computer & Digital Engineering

基　　金：国家自然科学基金项目(编号:51307074);江苏省自然科学基金项目(编号:BK20130466)资助。

摘　　要：为了提高微电网内部线路故障识别的实时性和准确性,建立适用性强的微电网故障诊断系统,文章提出了基于小波包能量熵和支持向量机(SVM)二叉树多分类器(2PTMC)的微电网内部线路故障诊断模型。利用小波包分析得到并网耦合点(PCC)处三相电流电压的小波包能量熵,作为样本的特征向量按故障发生频率训练多个SVM,构成树状微电网故障诊断系统。最后,在PSCAD平台建立微电网系统进行仿真,实验结果表明该方法可以准确地辨别出微电网故障类型,在未经训练的样本上也有良好的表现,符合微电网故障系统辨别的要求。In order to improve the real-time and accuracy of the line fault identification in the micro-grid,a micro-grid diagnosis system with strong applicability is established.In this paper,a micro-grid internal line fault diagnosis model named 2PTMC based on wavelet packet energy entropy and support vector machine is proposed.The wavelet packet energy entropy of the three-phase current and voltage at PCC is obtained by wavelet packet analysis.As the feature vector of the sample,multiple SVMs are trained according to the fault occurrence frequency to form a tree-like microgrid fault diagnosis system.Finally,the microgrid system is built on the PSCAD platform for simulation.The experimental results show that the method can accurately identify the type of microgrid faults and also has good performance on untrained samples,which meets the requirements of microgrid fault system identification.

关 键 词：小波包分析 支持向量机 二叉树 微电网 故障诊断 

分 类 号：TM712[电气工程—电力系统及自动化]