上海交通大学学报 ›› 2022, Vol. 56 ›› Issue (8): 1014-1023.doi: 10.16183/j.cnki.jsjtu.2021.195

• 新型电力系统与综合能源 • 上一篇    下一篇

基于多源局部放电信号数据流聚类分离方法

陈昌川1(), 刘凯1, 刘仁光1, 冯晓棕2, 覃延佳2, 代少升1, 张天骐1   

  1. 1.重庆邮电大学 通信与信息工程学院,重庆 400065
    2.广州友智电气技术有限公司,广州 510260
  • 收稿日期:2021-06-08 出版日期:2022-08-28 发布日期:2022-08-26
  • 作者简介:陈昌川(1978-),男,四川省广安市人,副教授,从事智能信息处理、图像人工智能处理、特高频局放检测、红外成像与测温研究。电话(Tel.):13350370998;E-mail: creditdegree@gmail.com.
  • 基金资助:
    国家自然科学基金面上项目(61671095);重庆市研究生教育教学改革研究重点项目(yjg192019);重庆市研究生教育教学改革研究一般项目(yjg213079);校企合作项目“电力设备局放在线监测系统装置”(SET20190627002)

Clustering Separation Method Based on Multi-Source Partial Discharge Signal Data Stream

CHEN Changchuan1(), LIU Kai1, LIU Renguang1, FENG Xiaozong2, QIN Yanjia2, DAI Shaosheng1, ZHANG Tianqi1   

  1. 1. School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China
    2. Guangzhou Youzhi Electric Technology Co., Ltd., Guangzhou 510260, China
  • Received:2021-06-08 Online:2022-08-28 Published:2022-08-26

摘要:

局部放电检测中, 多种放电源与现场干扰源同时存在且不断变化,导致多种局部放电源难以有效分离及识别.提出一种高效自适应在线数据流(EAOStream)聚类算法,该算法采用自然邻域创建K-dimensional树来提高查询近邻的效率,即通过流数据的特征得到自适应的邻域半径和区域密度,从而能够局部搜索并形成团簇,实现多种局部放电源的实时在线分离.在人工数据集和真实数据集验证了EAOStream的优越性,通过与传统的DenStream和SE-Stream算法比较,将其应用于气体绝缘变电站故障的模式识别.实验测试结果表明:EAOStream在真实的网络入侵检测、森林覆盖类型及多源局部放电信号数据集的聚类准确度分别达到95.28%、98.47%及97.23%,验证了该算法在气体绝缘变电站故障诊断方面的实用性和有效性.

关键词: 数据流, 聚类分离, 自适应, 自然邻域, 局部放电

Abstract:

In partial discharge(PD) detection, due to the simultaneous and constantly changing phenomenon of multiple discharge sources and on-site interference sources, it is difficult to effectively separate and identify multiple PD sources. An efficient adaptive efficient adaptive online data stream clustering algorithm (EAOStream) is proposed. The algorithm uses natural neighborhoods to create K-dimensional (KD) trees to improve the efficiency of querying neighbors. That is, the adaptive neighborhood radius and the area density are obtained through the characteristics of the flow data, which can search locally and form clusters, and realize the real-time online separation of multiple local discharge sources. The superiority of EAOStream is verified in the artificial data set and the real data set. After comparing EAOStream with the traditional DenStream and SE-Stream algorithms, it is applied to the pattern recognition of gas-insulated substation faults. Experimental test results show that the clustering accuracy of EAOStream in the real network intrusion detection, the forest cover type, and the multi-source PD signal data sets reaches 95.28%, 98.47%, and 97.23%, verifying the practicability and effectiveness of the algorithm in fault diagnosis of gas-insulated substations.

Key words: data stream, cluster separation, local adaptation, natural neighborhood, partial discharge (PD)

中图分类号: