电能质量扰动检测与识别方法综述  

A Review of Detection and Classification Methods for Power Quality Disturbances

作　　者：叶鹏 宋弘 吴浩 邱函 YE Peng;SONG Hong;WU Hao;QIU Han(School of Automation and Information Engineering,Sichuan University of Science and Engineering,Yibin 644000,China;Key Laboratory of Artificial Intelligence in Sichuan Province,Yibin 644000,China;Aba Teachers College,Aba 624000,China)

机构地区：[1]四川轻化工大学自动化与信息工程学院,四川宜宾644000 [2]人工智能四川省重点实验室,四川宜宾644000 [3]阿坝师范学院,四川阿坝624000

出　　处：《电工电气》2025年第2期1-9,共9页Electrotechnics Electric

基　　金：四川省科技厅项目(2022YFS0518)。

摘　　要：随着新能源发电设施的快速发展,电能质量扰动(PQDs)问题愈发严峻,对其高效检测与准确识别提出了更高要求。梳理了PQDs研究中包括信号特征检测精度不足、特征选择冗余及扰动类型识别能力有限等关键问题,对国内外相关研究成果进行归纳总结,详细阐述了电能质量扰动检测与识别方法的最新研究进展;探讨了基于先进信号处理技术的特征检测方法和智能算法的特征提取策略,以及依托深度学习模型的分类识别技术,分析了各类方法的优势与不足。指出在电能质量扰动检测与识别方面存在的问题,并对未来发展趋势进行了展望。With the rapid development of renewable energy generation facilities,the issue of power quality disturbances(PQDs)has become increasingly severe,raising higher demands for efficient detection and accurate identification.This paper first identifies key issues in PQDs research,including inadequate detection accuracy of signal characteristics,redundancy in feature selection,and limited capability in identifying types of disturbances.It summarizes relevant research findings from both domestic and international sources and elaborates on the latest advancements in detection and identification methods for power quality disturbances.Next,it focuses on feature detection methods based on advanced signal processing techniques,feature extraction strategies using intelligent algorithms,and classification and recognition techniques relying on deep learning models,offering a comprehensive analysis of the strengths and weaknesses of various approaches.Finally,the problems in power quality disturbance detection and identification are pointed out,and the future development trend is outlooked.

关 键 词：电能质量扰动 扰动检测 扰动信号 特征选择 扰动识别 深度学习模型 

分 类 号：TM712[电气工程—电力系统及自动化]