Fault Diagnosis Analysis of Power Transformer Based on PSO-BP Algorithm


Article Preview

BP neural network is currently the most widely used of neural network models in practical application in transformer fault diagnosis. BP algorithm is a local search algorithm which is easy to make the network into the local minimum values. Network training results are poor. It discusses PSO-BP algorithm which combines the particle swarm optimization (PSO) algorithm with the BP algorithm in this paper. It uses PSO algorithm to optimize the BP network’s weights and threshold. It is used in power transformer fault diagnosis. Experimental data results show that PSO-BP network fault diagnosis accuracy is higher than BP algorithm.



Advanced Materials Research (Volumes 466-467)

Edited by:

Wu Jinhui, Zhao Maotai and Wu Bo




H. Q. Sun et al., "Fault Diagnosis Analysis of Power Transformer Based on PSO-BP Algorithm", Advanced Materials Research, Vols. 466-467, pp. 789-793, 2012

Online since:

February 2012




[1] Sun Caixin, Chen Weigen , in: Technology of Online Monitoring Gas in Oil Of Electric Equipment And Fault Diagnosis. edtied by Press of Science, Beijing (2003) in press.

[2] Zhou Long: Transformer Fuzzy Diagnosis Research based on Neural Network. Transformer. Vol. 34 (1997), pp.32-35.

[3] RenXiaoBo, YangZhongXiu: Particle Swarm Optimization Algorithm Improved. Computer Engineering. Vol. 36 (2010), pp.205-207.

[4] Zhang Rui, Guo Ruijun: DGA Feature Extration based on the Chassification of Transformers Faults. High Voltage Technology. Vol. 31 (2005), pp.32-33.

[5] Qian Zheng, Yang Li and Yan Zhang: Selection of Typical Trainning Samples in Combined Neural Network Model. High Voltage Technology. Vol. 25 (1999), pp.1-6.