Fault Diagnosis of Cascaded Converter Based on Wavelet Packet RBF Neural Network

Abstract:

To study the power component open circuit faults diagnosis method of the cascaded converter. Aiming at the insufficiency of the BP learning algorithm in the machinery fault diagnosis, such as the low learning convergence speed, the easily appearing local minimum, the instability learning performance caused by the initial value, to proposed a new method applied to the cascaded converter based on radial basis function (RBF) neural network. Experiments show that the method based on wavelet packet analysis and RBF neural network has better learning and fault identification capability, and it can meet the online real-time fault diagnosis of the cascaded converter.

• Periodical: Applied Mechanics and Materials (Volume 273)
• Pages: 300-304
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.273.300
• Online since: January 2013
• Authors: Xin Wang, Juan Xu, Guo Dong Zhang, Rui Min Qi
• Keywords: Cascaded Converter, Fault Diagnosis, RBF Neural Network, Wavelet Packet Analysis

References

[1] Han Zhilingu, Ren Zhaohua, Qian Ming, 2006: Analysis of Open Circuit Malfunction for H Bridge Cascaded Multilevel High Voltage Converter, DianQi Chuandong, Vol. 31: 19-21.

[2] Yu Yong, Jiang Shengcheng, Yang Rongfeng, 2011: IGBT Open Circuit Fault Diagnosis Method for Inverter, Zhongguo Dianji Gongcheng Xuebao, Vol. 31: 30-35.

[3] Guo Fengyi, Ma Wenlong, Li Bin, 2010: Application of RBF neural network in fault diagnosis of circuit breaker, Transducer and Micro system Technologies, Vol. 29: 128-130.

[4] Luo Xiaoan, Weng Chenyu, 2007: Fault Diagnosis of High Voltage Circuit Breaker Based on RBF Neural Network, Dianqi Yingying, Vol. 26: 50-53.

[5] Wang Xuhong, He Yigang, 2010: Method of Fault Diagnosis for Induction Machine Rotor Broken Bar Based on Wavelet Packer and Elman Neural Network, Journal of Hunan University (Natural Science), Vol. 37: 45-48.