面向振动信号多特征值的电力变压器故障检测方法分析  

Analysis of power transformer fault detection based on multi-eigenvalue of vibration signal

作　　者：吴欣烨 WU Xinye(State Grid Shanghai Electric Power Company,Shanghai 200122,China)

机构地区：[1]国网上海市电力公司,上海200122

出　　处：《中国高新科技》2023年第15期85-87,共3页

摘　　要：对振动信号多特征值的电力变压器故障进行分析,要结合实际情况,明确电力变压器出现的直观问题。对某一领域的变压器或特定型号能够进行综合比较,明确各算法的应用范围。为了更好地解决上述问题,文章提出整理对应的电压等级或不同型号变压器,与典型故障下的振动信号完成比对分析。对100HC占比、谐波机电率等进行计算,二者之间的差异包含诊断故障差异性、流程差异性及截断正态分布拟合方法差异性。通过这种针对性的测量方法,最终的准确度可高达95%以上。适用于多特征值的电力变压器故障检测,保障其使用效果。To analyze the multi-eigenvalue fault of power transformer with vibration signal,it is necessary to make clear the problems of power transformer.The comprehensive comparison can be made to the transformer or specific model in a certain field,and the universal range of each algorithm can be defined.In order to better solve the above problems,this paper puts forward to collate the corresponding voltage grade or different types of transformers,and complete the analysis of vibration signals under typical faults.The difference between them includes the difference of fault diagnosis,the difference of process and the difference of fitting method of truncated normal distribution.Through this targeted measurement method,the final accuracy can be as high as 95%.It is suitable for multi-eigenvalue fault detection of power transformer to ensure its application effect.

关 键 词：振动信号 多特征值 电力变压器 故障检测分析 

分 类 号：TM41[电气工程—电器]