采用双向改进模糊2DLDA算法提升多因素影响的局部放电识别可靠性

Abstract
Figure/Table
References
Related Citation (15)

Download: PDF (769 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract The variation of defect size and applied voltage,and insulation aging can give rise to the dispersion and crossover of partial discharge (PD) features,which will influence the PD recognition reliability in transformers greatly. In order to solve the above problems,this paper presents a two-directional modified fuzzy 2DLDA (TD-MF2DLDA) algorithm,aiming to extract discriminatory features from PD gray images to infer the corresponding defect types. The method firstly designs a modified fuzzy 2DLDA (MF2DLDA) algorithm which is the foundation of TD-MF2DLDA for PD gray image compression. Finally,a fuzzy C-means (FCM) clustering tool is employed to carry out comparisons of different features. The recognition results of experimental PD samples in laboratory demonstrate that TD-MF2DLDA can achieve a successful clustering rate of 92.84% which is much higher than 77.80% of nonnegative matrix factorization aided principal component analysis (NMF-PCA) and 71.12% of traditional phase-resolved PD(PRPD) features. It could be found that TD-MF2DLDA can effectively eliminate the features dispersion and crossover derived from defect size,applied voltage and insulation aging,and thereby improve the PD recognition reliability. In addition,the FCM clustering validity measure Xie and Beni’s index (XBI) shows that features obtained by TD-MF2DLDA have better compactness within class and separability between classes than NMF-PCA and PRPD features,which is more suitable for on-site applications.

Key words： Transformers partial discharge feature extraction gray image modified two- directional fuzzy 2DLDA fuzzy C-means clustering

Received: 22 July 2013 Published: 22 January 2015

PACS:

TM835

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Wang Ke
	Liao Ruijin
	Wu Gaolin
	Wang Qian
	Wu Feifei

Cite this article:

Wang Ke,Liao Ruijin,Wu Gaolin等. Improvement of Partial Discharge Recognition Reliability Considering Influence of Multi-factors Based on Two- Directional Modified Fuzzy 2DLDA Algorithm[J]. Transactions of China Electrotechnical Society, 2014, 29(11): 210-220.

URL:

http://dgjsxb.ces-transaction.com/EN/ OR http://dgjsxb.ces-transaction.com/EN/Y2014/V29/I11/210

[1] 周凯,吴广宁,吴建东,等. 基于局部放电统计参量的脉冲电压下绝缘老化分析[J]. 电工技术学报,2008,23(4): 6-12. Zhou Kai,Wu Guangning,Wu Jiandong,et al. Analysis of aging properties for insulation based on statistical parameters of partial discharge under pulse voltage[J]. Transactions of China Electrotechnical Society,2008,23(4): 6-12.
[2] 张晓虹,张亮,乐波,等. 基于局部放电的矩特征分析大电机主绝缘的老化[J]. 中国电机工程学报,2002,22(5): 94-98. Zhang Xiaohong,Zhang Liang,Yue Bo,et al. Analysis on aging condition of stator winding insulation of generator based on the moment characteristics of partial discharge[J]. Proceedings of the CSEE,2002,22(5): 94-98.
[3] 高文胜,王猛,谈克雄,等. 油纸绝缘中局部放电的典型波形及其频谱特性[J]. 中国电机工程学报,2002,22(2): 1-5. Gao Wensheng,Wang Meng,Tan Kexiong,et al. Typical pulse shape and frequency characteristic of the PD signal in oil-paper insulation [J]. Proceedings of the CSEE,2002,22(2): 1-5.
[4] 杨丽君,廖瑞金,孙才新,等. 矢量量化在局部放电模式识别中的应用[J]. 中国电机工程学报,2009,29(31): 122-127. Yang Lijun,Liao Ruijin,Sun Caixin,et al. Application of vector quantization to partial discharge pattern recognition[J]. Proceedings of the CSEE,2009,29(31): 122-127.
[5] 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): 206-213. Wang Ke,Yang Lijun,Liao Ruijin,et al. Partial discharge pattern recognition based on discrete hidden Markov models[J]. Transactions of China Electrotech- nical Society,2011,26(8): 206-213.
[6] Li Jian,Jiang Tianyan,Harrison R F,et al. Recognition of ultra high frequency partial discharge signals using multi-scale features[J]. IEEE Transactions on Dielectrics and Electrical Insulation,2012,19(4): 1412-1430.
[7] Satish L,Zaengl W S. Can fractal features be used for recognizing 3-D partial discharge patterns[J]. IEEE Transactions on Dielectrics and Electrical Insulation,1995,2(3): 352-359.
[8] Krivda A,Gulski E,Satish L,et al. The use of fractal features for recognition of 3-D discharge patterns[J]. IEEE Transactions on Dielectrics and Electrical Insula- tion,1995,2(5): 889-892.
[9] Li Jian,Sun Caixin,Grzybowski S,et al. Partial discharge image recognition using a new group of features[J]. IEEE Transactions on Dielectrics and Electrical Insulation,2006,13(6): 1245-1253.
[10] Li Jian,Sun Caixin,Grzybowski S. Partial discharge image recognition influenced by fractal image compre- ssion[J]. IEEE Transactions on Dielectrics and Electrical Insulation,2008,15(2): 496-504.
[11] 唐炬,魏钢,李伟,等. 基于双向二维最大间距准则的局部放电灰度图像特征提取[J]. 电网技术,2011,35(3): 129-134. Tang Ju,Wei Gang,Li Wei,et al. Partial discharge gray image feature extraction based on bi-directional two-dimensional maximum margin criterion[J]. Power System Technology,2011,35(3): 129-134.
[12] Liao Ruijin,Wang Ke,Yang Lijun,et al. Partial discharge signal classification utilizing time-frequency representation and two directional 2DPCA[J]. Interna- tional Review of Electrical Engineering-IREE,2012,7(1): 3358-3369.
[13] Liao Ruijin,Wang Ke,Yang Lijun,et al. A new hybrid feature extraction method for partial discharge signals classification[J]. Pryzglad Elektrotechniczny,2012,88(11A): 191-195.
[14] Song Fengxi,Zhang David,Wang Jizhong,et al. A parameterized direct LDA and its application to face recognition[J]. Neurocomputing,2007,71(1-3): 191- 196.
[15] Moosavian A,Ahmadi H,Tabatabaeefar A. Fault diagnosis of main engine journal bearing based on vibration analysis using Fisher linear discriminant,K-nearest neighbor and support vector machine[J]. Journal of Vibroengineering,2012,14(2): 894-906.
[16] Calisir D,Dogantekin E. An automatic diabetes diagnosis system based on LDA-wavelet support vector Machine classifier[J]. Expert Systems with Applications,2011,38(7): 8311-8315.
[17] Zheng W S,Lai J H,Yuen P C. GA-Fisher: a new LDA-based face recognition algorithm with selection of principal components[J]. IEEE Transactions on Systems Man and Cybernetics Part B-Cybernetics,2005,35(5): 1065-1078.
[18] Xiong H L,Swamy M N S,Ahmad M O. Two- dimensional FLD for face recognition[J]. Pattern Recognition,2005,38(7): 1121-1124.
[19] 林宇生,房福龙,杨万扣. 模糊二维线性鉴别分析算法[J]. 无线电工程,2011,41(9): 15-17. Lin Yusheng,Fang Fulong,Yang Wankou. Fuzzy 2DLDA and its application to face recognition[J]. Radio Engineering,2011,41(9): 15-17.
[20] Yang W K,Yan X Y,Zhang L,et al. Feature extrac- tion based on fuzzy 2DLDA[J]. Neurocomputing,2010,73(10-12): 1556-1561.
[21] Li M,Yuan B Z. 2D-LDA: a statistical linear discri- minant analysis for image matrix[J]. Pattern Recognition Letters,2005,26(5): 527-532.
[22] Cavallini A,Chen X L,Montanari G C,et al. Diagnosis of EHV and HV transformers through an innovative partial-discharge-based technique[J]. IEEE Transactions on Power Delivery,2010,25(2): 814-824.
[23] Liao Ruijin,Yang Lijun,Li Jian,et al. Aging condition assessment of transformer oil-paper insulation model based on partial discharge analysis[J]. IEEE Transac- tions on Dielectrics and Electrical Insulation,2011,18(1): 303-311.
[24] 高新波. 模糊聚类分析及其应用[M]. 西安: 西安电子科技大学出版社,2004.
[25] Cavallini A,Montanari G C,Ciani F. Analysis of partial discharge phenomenon in paper-oil insulation systems as a basis for risk assessment evaluation[C]. Processing of IEEE International Conference on Dielectric Liquids,2005: 241-244.
[26] 杨丽君,孙才新,廖瑞金,等. 采用等效时频分析及模糊聚类法识别混合局部放电源[J]. 高电压技术,2010,36(7): 1710-1717. Yang Lijun,Sun Caixin,Liao Ruijin,et al. Applica- tion of equivalent time-frequency analysis and fuzzy clustering to recognize multiple PD sources[J]. High Voltage Engineering,2010,36(7): 1710-1717.
[27] 杨丽君,孙才新,廖瑞金,等. 油纸绝缘老化状态判别的局部放电特征量[J]. 电力系统自动化,2007,31(10): 55-60. Yang Lijun,Sun Caixin,Liao Ruijin,et al. Partial discharge features applied in aging condition discrimina- tion of oil-paper insulation[J]. Automation of Electric Power Systems,2007,31(10): 55-60.
[28] 杨丽君,廖瑞金,孙会刚,等. 油纸绝缘热老化特性及生成物的对比分析[J]. 中国电机工程学报,2008,28(22): 53-58. Yang Lijun,Liao Ruijin,Sun Huigang,et al. Contrasting analysis and investigation on properties and products of oil-paper during thermal aging process[J]. Procee- dings of the CSEE,2008,28(22): 53-58.
[29] Gulski E,Krivda A. Neural networks as a tool for recognition of partial discharges[J]. IEEE Transactions on Electrical Insulation,1993,28(6): 984-1001.
[30] James R E,Phung B T. Development of computer- based measurements and their application to PD pattern analysis[J]. IEEE Transactions on Dielectrics and Electrical Insulation,1995,2(5): 838-856.
[31] Candela R,Mirelli G,Schifani R. PD recognition by means of statistical and fractal parameters and a neural network[J]. IEEE Transactions on Dielectrics and Electrical Insulation,2000,7(1): 87-94.
[32] Wang Ke,Yang Lijun,Zhou Tianchun,et al. Study on thermal aging condition assessment of oil-paper insulation based on statistical features of partial discharge[C]. Proceedings of the IEEE 9th International Conference on the Properties and Applications of Dielectric Materials(ICPADM),2009: 325-328.
[33] Wang Ke,Liao Ruijin,Yang Lijun,et al. Nonnegative matrix factorization aided principal component analysis for high-resolution partial discharge image compression in transformers[J]. International Review of Electrical Engineering-IREE,2013,8(1): 479-490.