基于MPE和改进K⁃means算法的分接开关机械故障诊断方法  

Mechanical Fault Diagnosis of On⁃load Tap Changer Based on MPE and Improved K⁃means Algorithm

作　　者：马宏忠[1] 徐艳 魏海增 MA Hongzhong;XU Yan;WEI Haizeng(College of Energy and Electrical Engineering,Hohai University,Nanjing 211100,China;Chuzhou Power Supply Company,State Grid Anhui Electric Power Company,Anhui Chuzhou 239000,China)

机构地区：[1]河海大学能源与电气学院,南京211100 [2]国网安徽省电力有限公司滁州供电公司,安徽滁州239000

出　　处：《高压电器》2020年第8期198-204,共7页High Voltage Apparatus

基　　金：国家自然科学基金项目(51577050);国网江苏省电力有限公司2018年重点科技项目(J2018063)。

摘　　要：随着有载调压变压器在电网应用的增多以及有载分接开关(on⁃load tap⁃changer,OLTC)频繁地调节,分接开关的故障率在不断增加。为更有效进行OLTC机械故障诊断,提出一种基于MPE和改进K⁃means算法的OLTC机械故障诊断方法。首先,模拟OLTC的不同机械故障,采集振动信号;其次,为实现非线性振动信号下OLTC的故障诊断,采用多尺度排列熵(MPE)进行OLTC机械故障状态的特征提取;再次,采用粒子群(PSO)优化的K⁃means聚类算法诊断OLTC机械故障;最后,将该方法用于OLTC的机械故障诊断,并与传统K⁃means算法以及BP网络的诊断效果进行对比。结果表明,提出的基于MPE和改进K⁃means算法适用于OLTC机械故障诊断,诊断效果优于传统K⁃means算法以及BP网络,且其稳定性较高。A mechanical fault diagnosis method of on⁃load tap⁃changer(OLTC)is proposed based on MPE and im⁃proved K⁃means algorithm.First,different mechanical faults of OLTC are simulated,and the vibration signals are collected.Second,in order to realize the fault diagnosis of OLTC under nonlinear vibration signals,the multi⁃scale permutation entropy(MPE)is used to extract the features of OLTC mechanical fault state.Third,the K⁃means clus⁃tering algorithm optimized by particle swarm is used to diagnose the mechanical fault of OLTC.Finally,the proposed method is applied to the mechanical fault diagnosis of OLTC,and is compared with the methods based on traditional K⁃means algorithm and BP neural network.The results show that the proposed diagnosis method based on MPE and improved K⁃means algorithm is suitable for mechanical fault diagnosis of OLTC,and it has high stability and better diagnosis effect than the both methods based on traditional K⁃means algorithm and BP network.

关 键 词：OLTC K⁃means算法 粒子群优化的K⁃means算法 多尺度排列熵 

分 类 号：TM403.4[电气工程—电器]