基于经验模态分解及支持向量机的高压隔离开关机械故障诊断方法  

Mechanical Fault Diagnosis Method of High-voltage Disconnector Based on Empirical Modal Decomposition and Support Vector Machine

作　　者：郭煜敬 陈士刚 李少华 李洪涛 金光耀 张文涛 张一茗 关永刚[4] GUO Yujing;CHEN Shigang;LI Shaohua;LI Hongtao;JIN Guangyao;ZHANG Wentao;ZHANG Yiming;GUAN Yonggang(Pinggao Group Co.,Ltd.,Henan Pingdingshan 467001,China;School of Electrical Engineering,Beijing Jiaotong University,Beijing 100044,China;State Grid Jiangsu Electric Power Research Institute,Nanjing 211103,China;State Key Lab of Control and Simulation of Power Systems and Generation Equipments,Tsinghua University,Beijing 100084,China)

机构地区：[1]平高集团有限公司,河南平顶山467001 [2]北京交通大学电气工程学院,北京100044 [3]国网江苏省电力公司电力科学研究院,南京211103 [4]清华大学电机系电力系统及发电设备控制和仿真国家重点实验室,北京100084

出　　处：《高压电器》2018年第9期12-18,共7页High Voltage Apparatus

摘　　要：文中将K-means聚类算法和经验模态分解(empirical mode decomposition,EMD)相结合,对隔离开关机械故障进行诊断。为验证方法的有效性,搭建隔离开关运行状态在线监测系统,在某252 k V隔离开关的操动机构上选定位置安装了传感器,采集了机械振动等信号在模拟故障下的大量数据。首先利用小波包降噪方法对信号进行预处理;其次,应用EMD和谱分析方法对振动信号进行经验模态分解,得到IMF分量并将其能量熵作为特征量;然后,通过K-means聚类算法验证了特征提取方式的有效性;最后,通过支持向量机算法(support vector machine,SVM)对样本进行训练,实现了机械故障的准确诊断,验证了该方法的有效性。K-means clustering algorithm and EMD（empirlcal mode decomposition） are employed to diagnose mechanical fault of isolating switch. To verify the validity of the method, an on-line monitoring system of running state of disconnector is set up. A sensor is installed on the selected position of the actuating mechanism of a 252 kV isolating switch, and a large number of mechanical vibration data of the isolating switch with analog fault are collected. The signals are preprocessed firstly by wavelet packet noise reduction method, then EMD and spectral analysis method are applied for the mode decomposition of the vibration signals to obtain the IMF component, and the energy entropy of the IMF component is taken as the characteristic quantity. The effectiveness of the feature extraction method is verified by K-means clustering algorithm. Moreover, support vector maehine（SVM） is used to train the samples, hence accurate diagnosis of the mechanical faults is realized, and the validity of the proposed method is verified.

关 键 词：高压隔离开关 经验模态分解 能量熵 聚类 支持向量机 

分 类 号：TM564.1[电气工程—电器]