Envelope Signal of Partial Discharge Pattern Recognition Based on Wavelet Packet Transform

Abstract:

In order to achieve the GIS fault detection and defect type recognition, four typical defect models were designed and discharge tests are carried out aiming at insulation defect as well as discharge characteristics in the GIS .With a large number of ultra high frequency envelope signal ,a method of domain feature extraction was proposed based on wavelet packet transform with singular value decomposition .The envelope signal was decomposed through wavelet packet transform first in the method, then the coefficient matrix of wavelet packet transform was built in the scale ,after that feature vectors of matrix were extracted by means of singular value decomposition. On this basis, BP neural network was took advantage of for pattern recognition .The results show that the good recognition effect was obtained with that method . Keyword: Ultra high frequency; Envelope signal; Wavelet packet transform; Singular value decomposition; BP neural network

• Periodical: Advanced Materials Research (Volume 823)
• Pages: 536-540
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.823.536
• Online since: October 2013
• Authors: Dong Song Luo, Kun Peng Chen
• Keywords: BP Neural Network (BPNN), Envelope Signal, Singular Value Decomposition, Ultra-High Frequency (UHF), Wavelet Packet Transform

References

[1] Wang Changchang, Li Fuqi, Gao Shengyou. Online monitoring and fault diagnosis of power equipment [M]. Beijing: tsinghua university press, (2006).

[2] K. Raja. On the UHF partial discharge measurement in transformers [C]. in IEEE Electrical Insulation and Dielectric Phenomena, 2003, 349-352.

DOI: 10.1109/ceidp.2003.1254865

[3] Hans Gerd Kranz. PD pulse sequence analysis and relevance for on-site PD defect identification and evaluation[J]. IEEE Transactions on Dielectrics and Electrical Insulation, 2005, 12(2).

DOI: 10.1109/tdei.2005.1430397

[4] Zhang Yigang, Yu Wei-yong, HUANG Cheng-jun, etc. PD signal de-noising based on wavelet packet and Hidden Markov Models [J]. Journal of Shanghai jiaotong university, 2004 , 38(8).

[5] Jiang Qingyun, Zhang Chongyuan, Study on detection technology in the application of UHF PD signal extraction [J]. Journal of north China electric power university. 2006, 33(3): 10-13.

[6] Cheng Lizhi, Wang Hongxia. Wavelet theory and application [M]. Beijing: science press, (2004).

[7] Zhou Qian, Tang Ju, Tang Ming, et al. Mathematical model of four typical defects for UHF partialdischarge in GIS[J]. Proceedings of the CSEE, 2006, 26(8): 99-105.

[8] Zhang Defeng etc. MATLAB neural network application design [M] Beijing: mechanical industry press2009.