New Methold on Power Transformer Fault Diagnosis

Abstract:

This paper improves the simple genetic algorithm and combines genetic algorithm with BP algorithm to the wavelet neural network in the power transformer fault diagnosis by dissolved gas-in-oil analysis, Simulation result shows the problem was solved that wavelet network settles into local small extremum so easily that the network surging will increase and the network will not be convergent if the initialization is unreasonable, and overcomes the shortcoming that the speed is too slow if use genetic algorithm to train neural network independently.

• Periodical: Applied Mechanics and Materials (Volumes 217-219)
• Pages: 2623-2628
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.217-219.2623
• Online since: November 2012
• Authors: Nan Lan Wang, Ming Shan Cai
• Keywords: Fault Diagnosis, Genetic Algorithm (GA), Power Transformer, Wavelet Neural Network (WNN)

References

[1] GaoWen sheng, Gao Ning, Yan Zhang. Application of adaptive wavelet classification network for fault diagnosis in oil-immersed power equipment. Transactions of China Electrotechnical Society. 1998,13 (6):54-58.（In Chinese）

[2] Cai Chaohao. The early stage fault diagnosis of power transformer by artificial neural networks.Transformer. 1997, 34 (1): 34-37. （In Chinese）

[3] Yan Pingfan. Some views on the research of multilayer freeforward neural networks. ACTA AUTOMATICA SINICA).1997, 23 (1) :129-135. （In Chinese）

[4] WANG Shaofang, CAI Jinding, Liu Qingzhen. Power transformer fault diagnosis by improved hybrid algorithm and back propagation algorithm. Power System Technology. 2004, 28(4):30-33. （In Chinese）

[5] Li Jianmin, Lu Dunyi, Li Jianqing et al. Application of wavelet neural networks in fault diagnosis of transformer. Journal of Zhejiang University of Technology. 2001, 29 (1):56-59. （In Chinese）

[6] Zhang J, Walter G. Wavelet neural network for funtion learning. IEEE Trans on Signal Processing, 1995, 43(6):1485-1497. （In Chinese）

[7] Zhang Bangli, Li Yingguo, Cao Changxiu. Construction of wavelet neural network and its robustness analysis of learning algorithm. Journal of Chongqing University.1995, 18(6):88-95. （In Chinese）

[8] Li Hua,Sun Caixin,Hu Xuesong et al. Study of method on adaptive wavelets network for vibration fault diagnosis of steam turbine-generation set. Transactions of China Electrotechnical Society. 2000, 15 (3):57-60. （In Chinese）

[9] Zhao Xuezhi,Zou Chunhua,, Chen Tongjian et al. A research on the initialization of parameters of wavelet neural networks. Journal of South China University of Technology. 2003，31（2）：30-31（In Chinese）.

[10] Jiao Lichen. Application and realization of neural network. Xidian University Press, Xian.1993. （In Chinese）

[11] Xu Lina et al. neural network control, Publishing House of Electronics Industry, Beijing, 2003. （In Chinese）

[12] GUO Xia o ting, ZHU Yan.. Evolutionary neural networks based on genetic algorithms. Journal of Tsinghua University (Sci & Tech), 2000(4):116-119. （In Chinese）

[13] Zhou Jianhua. Transformer fault diagnosis research technology. Electrical engineering department of Southeast University, Nanjing 1998:1-19. （In Chinese）