Correlation Coefficient Matrix Based PD Feature Dimensional Reduction

Abstract:

The abstraction of diagnostic feature from field condition monitoring data is a significant research challenge. A new dimension reduction method based on correlation coefficient matrix is proposed aimed at the high-dimension characteristic parameters in the process of pattern recognition for partial discharge in power transformer. The CCM is constructed by parameters extracted from partial discharge signature in power transformer. The parameters that have similar classification characters are reduced directed by the correlation analysis result. The reduced PD features are inputted to the pattern classifiers of probabilistic neural networks (PNN). The results show that the parameter dimension is reduced and the classifier construction is simplified, and the recognition effect is better than that of the traditional back propagation neural network (BPNN) in the condition of small samples.

• Periodical: Advanced Materials Research (Volumes 945-949)
• Pages: 2499-2504
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.945-949.2499
• Online since: June 2014
• Authors: Yu Wang*, Jin Sha Yuan, Hai Kun Shang, Song Jin
• Keywords: Correlation Coefficient Matrix (CCM), Dimensional Reduction, Partial Discharge (PD), Pattern Recognition, Probabilistic Neural Networks

References

[1] Wang Guo-li, Hao Yan-peng, Li Yan-ming. Present and future development of partial discharge detection in power transformers[J]. Advanced Technology of Electrical Engineering and Energy, 2001, 20 (2): 52-57 (In Chinese).

[2] L. Satish and W.S. Zaengl. Can Fractal Features be used for recognizing 3-D partial discharge patterns, IEEE Transactions on Dielectrics and Electrical Insulation, 1995, Vol. 2, No. 3, pp.352-359.

DOI: 10.1109/94.395421

[3] Judd M D, Cleary G P, and Bennoch C J. Applying UHF partial discharge detection to power transformers [J]. IEEE Power Engineering Review, 2002, 22(8): 57-59.

DOI: 10.1109/mper.2002.4312468

[4] 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 (In Chinese).

[5] Amira A Mazroua. Discrimination between PD pulse shapes using different neural network paradigms [J]. IEEE Trans on Electrical Insulation, 1994, 1(6): 1119-1131.

DOI: 10.1109/94.368651

[6] Wang Yu, Yuan Jinsha and Jin Song, Partial Discharge Pattern Recognition Based on Gray Moment [J], Electrical Measurement & Instrumentation, Vol. 50, No. 574, pp.6-9, 2013 (In Chinese).

[7] E. Gulski and A. Krivda: Neural Networks as a Tool for Recognition of Partial Discharges, IEEE Transactions on Electrical Insulation, Vol. 28, No. 6, pp.984-1001, (1993).

DOI: 10.1109/14.249372

[8] N. C. Sahoo, M. M. A. Salama: Trends in Partial Discharge Pattern Classification: A Survey, IEEE Transactions on Dielectrics and Electrical Insulation Vol. 12, No. 2, p.248, (2005).

DOI: 10.1109/tdei.2005.1430395

[9] Wu Zhong-li, Yang Jian, Zhu Yong-li, et al. Power transformer fault diagnosis based on rough set theory and support vector machines [J]. Power System Protection and Control, 2010, 38(18): 80-83 (In Chinese).

[10] Liao Rui-jin, Yang Li-jun, Sun Cai-xin, 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.

[11] Wang Xiang-hong, ZHU Chang-ming, MAO Han-ling, et al. Source location of cracks in a turbine blade based on kernel principal component analysis and support vector machines [J]. Journal of Vibration and Shock, 2010, 29(11): 226-229 (In Chinese).

[12] Wang Guo-li, Hao Yan-peng, Li Yan-ming. Pulse current parameter revaluation of typical partial discharge models in oil [J]. Power System Technology, 2001, 25(10): 34-37 (In Chinese).

[13] Chang C S, Jin J, Chang C, et al. Separation of corona using wavelet packet transform and neural network for detection of partial discharge in gas-insulated substations[J]. IEEE Trans on Power Delivery, 2005, 20(2): 1363-1369.

DOI: 10.1109/tpwrd.2004.839187