Fault diagnosis of power electronic system based on fault gradation and neural network group

作者：Ma Chengcai; Gu Xiaodong^*; Wang Yuanyuan

来源：Neurocomputing, 2009, 72(13-15): 2909-2914.

DOI：10.1016/j.neucom.2008.06.033

• Fault diagnosis
• Neural network group
• Fault gradation

We propose a new fault diagnosis approach with fault gradation using BP (back-propagation) neural network group consisting of 3 sub BP neural networks. According to the hazard extents and the occurrence frequencies of different faults, the faults are divided into different grades. The higher the fault grade, the larger the number of the used sub neural networks is. Experimental results show that our approach makes the correctness rate of the fault diagnosis rise greatly (from less than 95.0% to 99.5%) and the performance of the whole fault diagnosis system gets much better especially for the online complex systems. The approach proposed in this paper also can be extended to other complex fault diagnosis systems, such as mechanical systems.

• 出版日期2009-8
• 单位复旦大学