基于奇异值分解的局部放电模式识别方法

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (6157 KB)
输出: BibTeX | EndNote (RIS)

摘要提出了一种基于奇异值分解的变压器局部放电模式识别方法。通过搭建人工缺陷实验环境并采集样本数据,计算每个样本的统计特征参数,构成实验数据的样本矩阵。对样本矩阵进行奇异值分解,判断保留矩阵的特征是否明显,确定最佳保留矩阵的阶数,从而得到降维后的类型特征空间描述矩阵和类中心描述向量组。对现场采集的样本数据进行计算得到待分类的样本向量,并用类型特征空间描述矩阵进行线性变换,然后计算变换后的向量与类中心向量组中每个向量的距离,从而得到分类的判断结果。该算法简单而且高效,能够实现局部放电检测中各种放电信号的有效区分,局部放电模式识别召回率约为91.3%。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	阮羚
	李成华
	宿磊
	谢齐家
	吴玉佳
	张新访

关键词 ：变压器, 局部放电, 奇异值分解, 模式识别

Abstract：A pattern recognition method based on singular value decomposition (SVD) for partial discharge in transformers is proposed. By setting up an experimental environment with artificial defects and calculating the statistical parameters from the data obtained from each sample, the sample matrix is constructed. SVD is then carried out for the sample matrix. After dimensional reduction by decomposing the matrix, the best order for the remained matrix is judged by the singular value. Then, the low-dimensional description matrix of feature space and the class-center vectors are obtained. The classified sample vector which is acquired on Site is formulated by linear transforms of the description matrix. The result of classification is gotten by calculating the distances between the transformed vector and the class-center vector. The proposed method is simple and efficient. It has the ability to recognize effectively various signals of partial discharge. The experiments show that the recall rate of partial discharge is about 91.3%.

Key words： Transformer partial discharge singular value decomposition pattern recognition

收稿日期: 2013-06-30 出版日期: 2015-10-20

PACS:

TM83

作者简介: 阮羚男,1961年生,教授级高工,研究方向为高电压工程和电气绝缘。李成华男,1972年生,博士,副教授,硕士生导师,研究方向为数据挖掘和云计算技术。

引用本文:

阮羚, 李成华, 宿磊, 谢齐家, 吴玉佳, 张新访. 基于奇异值分解的局部放电模式识别方法[J]. 电工技术学报, 2015, 30(18): 223-228. Ruan Ling, Li Chenghua, Su Lei, Xie Qijia, Wu Yujia, Zhang Xinfang. Pattern Recognition for Partial Discharging Using Singular Value Decomposition. Transactions of China Electrotechnical Society, 2015, 30(18): 223-228.

链接本文:

https://dgjsxb.ces-transaction.com/CN/Y2015/V30/I18/223

[1] 汪可, 杨丽君, 廖瑞金, 等. 基于离散隐式马尔科夫模型的局部放电模式识别[J]. 电工技术学报, 2011, 26(8): 205-212. Wang Ke, Yang Lijun, Liao Ruijin, et al. Partial discharge pattern recognition based on discrete hidden Markov models[J]. Transactions of China Electrotechnical Society, 2011, 26(8): 205-212.
[2] 姚林朋, 郑文栋, 钱勇, 等. 基于AdaBoost的局部放电综合特征决策树识别方法[J]. 电力系统保护与控制, 2011, 39(21): 105-109. Yao Linpeng, Zheng Wendong, Qian Yong, et al. Pattern recognition based on AdaBoost decision tree for partial discharge[J]. Power System Protection and Control, 2011, 39(21): 105-109.
[3] 任先文, 薛雷, 宋阳, 等. 基于分形特征的最小二乘支持向量机局部放电模式识别[J]. 电力系统保护与控制, 2011, 39(14): 143-147. Ren Xianwen, Xue Lei, Song Yang, et al. The pattern recognition of partial discharge based on fractal characteristics using LS-SVM[J]. Power System Protection and Control, 2011, 39(14): 143-147.
[4] 唐炬, 王静, 李剑, 等. 统计参数用于局部放电模式识别的研究[J]. 高电压技术, 2002, 28(8): 4-6. Tang Ju, Wang Jing, Li Jian, et al. Statistical parameter method for PD pattern recognition[J]. High Voltage Engineering, 2002, 28(8): 4-6.
[5] 胡文堂, 高胜友, 余绍峰, 等. 统计参数在变压器局部放电模式识别中的应用[J]. 高电压技术, 2009, 35(2): 277-281. Hu Wentang, Gao Shengyou, Yu Shaofeng, et al. Application of statistic parameters in recognition of partial discharge in transformers[J]. High Voltage Engineering, 2009, 35(2): 277-281.
[6] 廖瑞金, 杨丽君, 孙才新, 等. 基于局部放电主成分因子向量的油纸绝缘老化状态统计分析[J]. 中国电机工程学报, 2006, 26(14): 114-119. Liao Ruijin, Yang Lijun, Sun Caixin, et al. Aging condition assessment of oil-paper based on principal component and factor analysis of partial discharge[J]. Proceedings of the CSEE, 2006, 26(14): 114-119.
[7] 周天春, 杨丽君, 廖瑞金, 等. 基于局部放电因子向量和BP神经网络的油纸绝缘老化状况诊断[J]. 电工技术学报, 2010, 25(10): 18-23. Zhou Tianchun, Yang Lijun, Liao Ruijin, et al. Diagnosis of aging condition in oil-paper insulation based on factor vectors of partial discharge and BP neural network[J]. Transactions of China Electro- technical Society, 2010, 25(10): 18-23.
[8] 王海跃, 王国利. 自适应遗传算法在变压器超高频局放模式识别中的应用[J]. 湖南电力, 2004, 24(5): 4-7. Wang Haiyue, Wang Guoli. Application of application of adaptive genetic algorithm to transformer ultra-high- frequency PD pattern recognition[J]. Hunan Electric Power, 2004, 24(5): 4-7.
[9] 徐志向, 侯世英, 周林, 等. 基于奇异值分解的电力系统谐波状态估计[J]. 电力自动化设备, 2006, 26(11): 28-31. Xu Zhixiang, Hou Shiying, Zhou Lin, et al. Power system harmonic state estimation based on singular value decomposition[J]. Electric Power Automation Equipment, 2006, 26(11): 28-31.
[10] 李天云, 陈昌雷, 周博, 等. 奇异值分解和最小二乘支持向量机在电能质量扰动识别中的应用[J]. 中国电机工程学报, 2008, 28(34): 124-128. Li Tianyun, Chen Changlei, Zhou Bo, et al. Application of SVD and LS-SVM in power quality disturbances classification[J]. Proceedings of the CSEE, 2008, 28(34): 124-128.
[11] 高彩亮, 岳苓, 黄少先. 一种基于小波分析和奇异值分解的故障选相方法[C]. 中国高等学校电力系统及其自动化专业第二十五届学术年会论文集, 长沙, 2009.
[12] 张琳, 曹一家. 基于奇异值分解方法的FACTS交互影响分析[J]. 电力系统自动化, 2008, 32(5): 20-24. Zhang Lin, Cao Yijia. Analysis on the interaction of FACTS controllers based on the SVD method[J]. Automation of Electric Power Systems, 2008, 32(5): 20-24.
[13] 唐炬, 李伟, 姚陈果, 等. 局部放电干扰评价参数信噪比的二阶估计[J]. 中国电机工程学报, 2011, 31(7): 125-130. Tang Ju, Li Wei, Yao Chenguo, et al. Two-order estimation of interference evaluation parameter SNR on partial discharge[J]. Proceedings of the CSEE, 2011, 31(7): 125-130.
[14] 杜林, 戴斌, 陆国俊, 等. 基于S变换局部奇异值分解的过电压特征提取[J]. 电工技术学报, 2010, 25(12): 147-153. Du Lin, Dai Bin, Lu Guojun, et al. Overvoltage features extraction based on S transform and local singular value decomposition[J]. Transactions of China Electrotechnical Society, 2010, 25(12): 147-153.
[15] 胡卫红, 舒泓, 栾宇光. 基于奇异值分解的电能质量信号去噪[J]. 电力系统保护与控制, 2010, 38(2): 31-33 Hu Weihong, Shu Hong, Luan Yuguang. Power quality signals’ de-noising method based on singular value decomposition (SVD)[J]. Power System Protection and Control, 2010, 38(2): 31-33.
[16] 唐炬, 李伟, 欧阳有鹏. 采用小波变换奇异值分解方法的局部放电模式识别[J]. 高电压技术, 2010, 36(7): 1686-1691. Tang Ju, Li Wei, Ouyang Youpeng. Partial discharge pattern recognition using discrete wavelet transform and singular value decomposition[J]. High Voltage Engineering, 2010, 36(7): 1686-1691.