油浸式变压器故障诊断方法研究  

Research on Fault Diagnosis Methods for Oil Immersed Transformers

作　　者：王元彬 王亮 王鹏 席凌青 曾利 WANG Yuanbin;WANG Liang;WANG Peng;XI Lingqing;ZENG Li(Gongzui Hydropower Station,CHN Energy Dadu River Co.,Ltd.,Leshan 614900,China)

机构地区：[1]国能大渡河公司龚嘴水力发电总厂,四川乐山614900

出　　处：《电工技术》2024年第24期238-242,共5页Electric Engineering

摘　　要：变压器是电力系统中不可或缺的重要设备,因此基于变压器的故障诊断技术尤为重要。首先,介绍了油浸式变压器的故障类型,分为外部故障、内部故障;然后,阐述了DGA气体分析、油质分析、热影像检测和局部放电检测等传统故障诊断方法的应用范围和优缺点;最后,对智能诊断技术模糊理论、神经网络、SVM及各改进方法进行总体概括归纳,分析不同故障诊断方法在油浸式变压器诊断中的不同效果。具体归纳了变压器传统诊断方法和智能诊断方法的优缺点,为变压器故障诊断的新技术提供了更多的可能。Transformers are indispensable and important equipment in the power system,so transformer based fault diagnosis technology is particularly important.Firstly,this article introduces the types of faults in oil immersed transformers,which can be divided into external faults and internal faults.Then,the application scope and advantages and disadvantages of traditional fault diagnosis methods such as DGA gas analysis,oil quality analysis,thermal image detection,and partial discharge detection were elaborated.Finally,an overall summary and summary of intelligent diagnostic techniques such as fuzzy theory,neural networks,SVM,and various improvement methods are presented,and the different effects of different fault diagnosis methods in the diagnosis of oil immersed transformers are analyzed.This article specifically summarizes the advantages and disadvantages of traditional and intelligent diagnostic methods for transformers,providing more possibilities for new technologies in transformer fault diagnosis.

关 键 词：油浸式变压器 故障诊断 DGA 机器学习 

分 类 号：TM407[电气工程—电器]