基于VMD与优化SVM的变压器绕组松动缺陷振动信号诊断方法  

Vibration Signal Diagnosis Method of Transformer Winding Looseness Defect Based on VMD and Optimized SVM

作　　者：顾仲翔 马宏忠[1] 张勇 陈冰冰 李勇 GU Zhongxiang;MA Hongzhong;ZHANG Yong;CHEN Bingbing;LI Yong(College of Energy and Electrical Engineering,Hohai University,Nanjing 211100,China;State Grid Jiangsu Electric Power Company Nanjing Power Supply Company,Nanjing 210008,China)

机构地区：[1]河海大学能源与电气学院,南京211100 [2]国网江苏省电力有限公司南京供电分公司,南京210008

出　　处：《高压电器》2023年第1期117-125,共9页High Voltage Apparatus

基　　金：国网江苏省电力公司重点科技项目(J2020042);111引智计划(B14022)。

摘　　要：随着电力系统中变压器容量的不断增加,变压器绕组松动缺陷引起的影响也愈发严重,故需进行故障诊断。针对利用振动信号进行变压器绕组松动缺陷诊断问题,提出基于变分模态分解(VMD)排列熵(PE)的变压器振动信号特征提取方法与天牛须搜索(BAS)优化支持向量机(SVM)的变压器绕组松动缺陷诊断方法。首先对一台实际110 kV变压器设置不同松动状态,采集绕组正常与不同松动程度状态下振动信号;其次,采用变分模态分解结合排列熵进行变压器绕组松动缺陷特征提取;再次,采用天牛须搜索优化支持向量机算法进行绕组松动状态模式识别。最后将该方法与传统方法进行对比,结果表明,VMD分解效果优于当前主要采用的经验模态分解,排列熵量化故障特征效果优于多尺度排列熵、近似熵等时间序列复杂度计算指标,BAS⁃SVM运算时间及诊断正确率优于网格搜索优化SVM、遗传算法优化SVM等优化算法;所提方法在所进行的60%~110%额定电流状态试验下变压器绕组松动故障诊断准确率达到98.7%以上。With the continuous increase of transformer capacity in power system,the influence due to looseness de⁃fect of transformer winding is becoming more and more serious,and it is therefore necessary to perform fault diagno⁃sis.In view of looseness defect diagnosis of transformer winding by the use of vibration signal,the feature extraction method of transformer vibration signal based on variational mode decomposition(VMD)permutation entropy(PE)and transformer winding looseness defect diagnosis method based on support vector machine(SVM)optimized by bee⁃tle antennae search(BAS)are proposed.Firstly,one set of practical 110 kV transformer is set with different loose⁃ness states to collect vibration signals under normal and different degrees of looseness;Then,the variational mode decomposition combined with permutation entropy is used for feature extraction of looseness defects of transformer winding.After that,the support vector machine optimized by beetle antennae search algorithm is used for pattern rec⁃ognition of winding looseness.Finally,the method is compared with the the traditional method.The results show that VMD decomposition effect is better than presently mainly⁃used empirical mode decomposition,and the permutation entropy fault feature effect is better than the time series complexity calculation index of multi⁃scale permutation en⁃tropy and approximate entropy.The BAS⁃SVM operation time and diagnosis accuracy rate are better than such optimi⁃zation algorithms as grid search optimization SVM and genetic algorithm optimization SVM.The accuracy of the fault diagnosis of the transformer winding looseness tested at 60%~110%rated current by the proposed method is above 98.7%.

关 键 词：变压器 绕组松动 振动信号 变分模态分解 排列熵 天牛须搜索 

分 类 号：TM407[电气工程—电器] TP18[自动化与计算机技术—控制理论与控制工程]