变压器状态评估及故障诊断研究综述  

Review of Transformer ConditionAssessment and Fault Diagnosis

作　　者：梁栋 朱建华 张翠 康诗奇 LIANG Dong;ZHU Jianhua;ZHANG Cui;KANG Shiqi(Xi'an Xibian Components Co.,Ltd.,Xi'an 710077,China;Xi'an Jiaotong University,Xi'an 710049,China)

机构地区：[1]西安西变组件有限公司,陕西西安710077 [2]西安交通大学,陕西西安710049

出　　处：《变压器》2024年第2期35-43,共9页Transformer

摘　　要：电力变压器状态评估及故障诊断为设备安全稳定运行提供了重要保障。在电力大数据广泛应用的背景下,智能电网结构快速构建,电力设备状态数据呈现出数量大、类型多等特征,因而变压器状态评估及故障诊断算法由阈值判断法逐步过渡为机器学习等算法。本文作者总结了近年来国内外变压器监测研究中采用的方法;概述了变压器状态评估和故障诊断领域的研究现状,介绍了常用算法相关原理,包括模糊理论法、集对分析法、传统机器学习算法、预测算法和深度机器学习算法等;分析了目前该领域亟需解决的问题,并对未来研究方向进行了展望。Power transformer condition assessment and fault diagnosis are of great importance to ensure the safety and stable operation of equipment.With wide application of power big data,the smart grid structure is rapidly constructed,and the condition data of power equipment has shown more characteristics of large quantity and multiple types.Therefore,transformer condition assessment and fault diagnosis algorithm has gradually transitioned from threshold judgment method to machine learning method.In this paper,the methods adopted in transformer monitoring studies from home and abroad in recent years are summarized,and the research status of transformer condition assessment and fault diagnosis is outlined.Besides,the principles of several common algorithms are introduced,including fuzzy theory,set-pair analysis,traditional machine learning,predictive algorithms and deep learning.On this basis,the technical problems faced in the field of transformer condition assessment and fault diagnosis are analyzed,and the main research directions in the future is prospected.

关 键 词：电力变压器 人工智能 状态监测 状态评估 故障诊断 

分 类 号：TM411[电气工程—电器]