基于小波树的电容型设备异常检测方法

吉林大学学报(理学版)

基于小波树的电容型设备异常检测方法

虞强源^1,2, 范世超^1, 金承业³

1. 吉林大学计算机科学与技术学院, 长春 130012； 2. 吉林大学符号计算与知识工程教育部重点实验室, 长春 130012；3. 吉林省乐金电子有限公司, 长春 130062

收稿日期:2014-10-27 出版日期:2015-11-26 发布日期:2015-11-23
通讯作者: 虞强源 E-mail:qiangyuan@jlu.edu.cn

Abnormality Detection of Capacitive Equipment Based on Wavelet Tree

YU Qiangyuan^1,2, FAN Shichao^1, JIN Chengye³

1. College of Computer Science and Technology, Jilin University, Changchun 130012, China;2. Key Laboratory of Symbolic Computation and Knowledge
Engineering of Ministry of Education, [JP+1]Jilin University, Changchun 130012, China; 3. Jilin Province Lejin Electronic Co. Ltd., Changchun 130062, China

Received:2014-10-27 Online:2015-11-26 Published:2015-11-23
Contact: YU Qiangyuan E-mail:qiangyuan@jlu.edu.cn

摘要/Abstract

摘要：

针对电力系统中电容型设备的异常检测问题, 以介质损耗因数为主要监测参数, 提出一种基于小波树的设备绝缘性异常检测方法, 以解决异常的自动检测问题. 该方法利用在线监测获得的介质损耗因数数据, 采用二层小波树方法进行异常检测, 通过检测介质损耗因数的突发性和持续性突变异常, 分析得到设备的绝缘性异常检测结果. 根据对变电所电容型设备实测数据的分析, 验证了该方法的有效性和实用性, 为设备的异常预警提供了辅助支持.

关键词: 异常检测, 小波树, 电容型设备, 介质损耗因数

Abstract:

To solve the abnormality detection problem of capacitive equipment in the power system, a method of electric insulativity detection based on wavelet tree was proposed with dielectric loss factor as detecting parameter. It basically solves the problem of abnormality automatic detection. In this method, twolayered wavelet tree was used to deal with the data of dielectric loss factor, which can detect abrupt change and continuous change of dielectric loss factor. The real dielectric loss factor data obtained from power substation were used to analyze the electric insulativity of capacitive equipment. The result proves the effectivity and practicality of the method, which can assist the early warning of abnormality.

Key words: abnormality detection, wavelet tree, capacitive equipment, dielectric loss factor

中图分类号:

TP274

虞强源, 范世超, 金承业. 基于小波树的电容型设备异常检测方法[J]. 吉林大学学报(理学版), 2015, 53(06): 1275-1281.

YU Qiangyuan, FAN Shichao, JIN Chengye. Abnormality Detection of Capacitive Equipment Based on Wavelet Tree[J]. Journal of Jilin University Science Edition, 2015, 53(06): 1275-1281.

[1]	姚艳秋. 基于趋势分析的变电站设备异常检测方法[J]. 吉林大学学报(理学版), 2021, 59(3): 649-652.
[2]	吴玲云, 秦贵和, 于赫. 基于随机森林的车载CAN总线异常检测方法[J]. 吉林大学学报(理学版), 2018, 56(3): 663-668.
[3]	于潇菂, 努尔布力, 胡亮, 解男男. 一种基于数据挖掘的多步入侵警报关联模型[J]. 吉林大学学报(理学版), 2013, 51(05): 881-886.
[4]	徐智博, 李宝华, 金承业. 基于离散傅里叶变换计算介质损耗因数的二周期修正法[J]. J4, 2012, 50(02): 328-332.
[5]	胡亮, 金刚, 于漫, 任斐, 任维武. 基于异常检测的入侵检测技术[J]. J4, 2009, 47(6): 1264-1270.
[6]	胡亮, 任维武, 任斐, 刘晓博, 金刚. 基于改进密度聚类的异常检测算法[J]. J4, 2009, 47(05): 954-960.