Application of Multi-Sensor Information Fusion Technology in the Fault Diagnosis of Transformer

Abstract:

The information fusion method is introduced into the transformer fault diagnosis. Through the sensor acquire transformer in operation of each state parameter, using two parallel BP neural networks to local diagnosis, with D-S evidence theory to global fuse the local diagnostic results. It realized the accurate diagnosis when transformer comes out one or a variety of faults at the same time. The experiments demonstrate that the credibility of diagnosis results are improved significantly, uncertainties are obviously reduced, which fully shows that the method is effective.

• Periodical: Applied Mechanics and Materials (Volumes 385-386)
• Pages: 601-604
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.385-386.601
• Online since: August 2013
• Authors: Han Min Ye, Zun Ding Xiao
• Keywords: BP Neural Network (BPNN), D-S Evidence Theory, Fault Diagnosis, Transformer

References

[1] Chen Xingang, Tian Xiaoxiao. A method of transformer fault diagnosis in combination with DS evidence information fusion[J]. Transformer. 2010, 47 (8): 59-63.

[2] Jian Xiaogang, Jia Hongsheng. Advance on multi-sensor information fusion technologies[J]. Chinese Journal of Construction Machinery. 2009, 7 (2): 227-232.

[3] Liao Ruijin, Liao Yuxiang. Study on Synthetic Diagnosis Method of Transformer Fault fusing Mulit-neural Network and Evidence theory [J]. Proceedings of the CSEE, 2006, 26(3): 119-123.

[4] Yang Haia, Liu Jin. Application of BP Neural network to Transformer fault diagnosis [J]. Transformer, 2009, 46(1): 67-70.

[5] Zhu Hanqing, Ma Zhenshu. Fault Diagnosis Method of Data Fusion based on Neural Network and D- S Evidence Theory[J]. Mechanical vibration. 2012, 36(10): 90-93.