|本期目录/Table of Contents|

[1]汪可,张书琦,李金忠,等. 基于灰度图像分解的局部放电特征提取与优化[J].电机与控制学报,2018,22(05):25-34.[doi:10.15938/j.emc.2018.05.004]
 WANG Ke,ZHANG Shu-qi,LI Jin-zhong,et al. Partial discharge feature extraction and optimizationbased on gray image decomposition[J].,2018,22(05):25-34.[doi:10.15938/j.emc.2018.05.004]
点击复制

 基于灰度图像分解的局部放电特征提取与优化(PDF)
分享到:

《电机与控制学报》[ISSN:1007-449X/CN:23-1 408/TM]

卷:
22
期数:
2018年05
页码:
25-34
栏目:
出版日期:
2018-05-15

文章信息/Info

Title:
 Partial discharge feature extraction and optimizationbased on gray image decomposition
作者:
 汪可1 张书琦1 李金忠1 孙建涛1 赵晓宇1 廖瑞金2 邹国平3
 (1. 中国电力科学研究院,北京 100192;2. 重庆大学 输配电装备及系统安全与新技术国家重点实验室,重庆 400044;3. 国网浙江省电力公司电力科学研究院,浙江 杭州 310014)
Author(s):
  WANG Ke1 ZHANG Shu-qi1 LI Jin-zhong1 SUN Jian-tao1 ZHAO Xiao-yu1LIAO Rui-jin2 ZOU Guo-ping3
 (1. China Electric Power Research Institute,Beijing 100192,China;2. State Key Laboratory of Power TransmissionEquipment & System Security and New Technology,Chongqing University,Chongqing 400044,China;3. State Grid Zhejiang Electric Power Research InstituteHangzhou 310014,China)
关键词:
局部放电 模式识别 图像分解 特征提取 特征选择 模糊 k 近邻
Keywords:
partial discharge pattern recognition image decomposition feature extraction feature selec-tion fuzzy k-nearest neighbor
分类号:
-
DOI:
10.15938/j.emc.2018.05.004
文献标志码:
A
摘要:
 提取有效的局部放电( PD) 特征是输变电设备缺陷识别的前提。以局部放电灰度图像为分析对象,提出了基于二维主成分分析( 2DPCA) 的局部放电图像特征提取策略。算法通过 2DPCA将 PD 灰度图像分解为多个一维向量,并对每个向量提取了 9 个特征参量,组成了 PD 图像分解特征集。同时,建立了基于粒子群优化( PSO) 算法的 PD 特征选择算法,以优化 PD 图像分解特征,提升局部放电缺陷类型识别结果。对实验室考虑多因素影响的 PD 样本识别结果表明,2DPCA 图像分解特征可以取得 93% 的 PD 缺陷识别率,经过 PSO 优化后的 2DPCA 特征可以将 PD 识别率提高至 96% ,并且特征维数由 72 降至 28,充分说明方法的有效性。另外,对添加不同随机干扰的 PD 样本平均识别率均大于 85% ,表明 2DPCA 图像特征具有较好的抗随机干扰能力。
Abstract:
 Effective features extraction of partial discharge ( PD) is the foundation of defect identificationof electrical apparatus. Using PD gray image as the analysis object,
a PD image features extraction strate-gy was proposed based on two-dimensional principal component analysis ( 2DPCA) . Various 1-dimen-sional (1D) vectors were obtained by implementing 2DPCA on PD gray images in the proposed method.9 characteristic parameters were extracted from each 1D vector,
which constituted the PD image decompo-sition features. In addition,a PD features selection algorithm was developed based on particle swarm op-timization ( PSO) algorithm,which attempts to optimize the extracted PD image decomposition featuresand improve the PD recognition accuracy. The recognition results of PD samples considering the multi-actor influences in laboratory illustrate that the proposed 2DPCA image decomposition features can a-chieve the high PD recognition accuracy of 93% . Besides,the PSO optimized 2DPCA features can fur-ther improve the PD recognition accuracy to 96% and simultaneously reduce the feature dimension from72 to 28,which fully demonstrates effectiveness of the proposed algorithm. Moreover,the average recog-nition accuracies of PD samples added with different random noises are all higher than 85% ,which indi-cates that 2DPCA image features possess good tolerance ability of random noises.

参考文献/References:

-

相似文献/References:

[1]杨永明,陈学军,陈民铀,等.电机局部放电超声特性[J].电机与控制学报,2011,(10):63.
 YANG Yong-ming,CHEN Xue-jun,CHEN Min-you,et al.Ultrasonic characteristics of motor partial discharge[J].,2011,(05):63.
[2]陈庆国,蒲金雨,丁继媛,等. 电力电缆局部放电的高频与特高频联合检测[J].电机与控制学报,2013,(04):39.
 CHEN Qing-guo,PU Jin-yu,DING Ji-yuan,et al.[J].,2013,(05):39.
[3]彭超,阮江军,黄道春,等. 基于特高频谱图统计参量的局部放电定位方法研究[J].电机与控制学报,2014,18(09):1.
 PENG Chao,UAN Jiang-jun,HUANG Dao-chun,et al. Research on partial discharge location method based onPRPD statistic parameters of ultra-high frequency[J].,2014,18(05):1.
[4]尤波,李忠杰,黄玲,等.手部抓取动作特征提取算法研究[J].电机与控制学报,2017,21(12):75.[doi:10.15938/j.emc.2017.12.010]
 YOU Bo,LI Zhong-jie,HUANG Ling,et al.Feature extraction algorithm for hand motion modes[J].,2017,21(05):75.[doi:10.15938/j.emc.2017.12.010]

备注/Memo

备注/Memo:
 收稿日期: 2016 - 03 - 12
基金项目: 国家电网公司科技项目( 5211DS16000G)
作者简介: 汪 可( 1987—) ,男,博士,高级工程师,研究方向为高电压与绝缘技术及变压器相关技术;
张书琦( 1981—) ,男,硕士,高级工程师,研究方向为高电压与绝缘技术及变压器相关技术;
李金忠( 1974—) ,男,博士,教授级高级工程师,研究方向为高电压与绝缘技术及变压器相关技术;
孙建涛( 1980—) ,男,博士,高级工程师,研究方向为高电压与绝缘技术及变压器相关技术;
赵晓宇( 1987—) ,男,硕士,工程师,研究方向为高电压与绝缘技术及变压器相关技术;
廖瑞金( 1963—) ,男,博士,教授,博士生导师,研究方向为电气设备绝缘在线监测与故障诊断研究和高电压测试技术;
邹国平( 1982—) ,男,博士,高级工程师,研究方向为电磁暂态分析、输变电设备状态检修和带电检测技术。
更新日期/Last Update: 2018-07-02