Fault Diagnosis Technology for Power Electronics Rectifier Based on Wavelet Packet

Abstract:

The paper presents on power electronics controlled rectifier faults diagnose technology based on the combination wavelet packet transformation and neural network mainly. By using the character of wavelet packet multiresolution, fault signal is decomposed at multi-scale, orthogonalation and normalization, extract feature vector, which as training input of neural network, and design classifier of fault pattern. The validity and feasibility of the fault diagnosis method is demonstrated by simulation. this method can quite accurately diagnosis fault and define fault element for power electronics controlled rectifier, and the diagnosis precision is high, the method has very good practical value and apply future on solving power electronics controlled rectifier fault diagnosis.

• Periodical: Advanced Materials Research (Volumes 317-319)
• Pages: 1215-1218
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.317-319.1215
• Online since: August 2011
• Authors: Qing Hu, Rong Jie Wang, Lian Shi Lin
• Keywords: BP Neural Network (BPNN), Fault Diagnosis, Power Electronics, Wavelet Packet (WP)

References

[1] Diqi Zhu, Shenlin Yu. Neural networks data fusion algorithm of electronic equipment fault diagnosis. Proc.5th World Congress on Intelligent Control and Automation, pp:1815~1818,2004.

DOI: 10.1109/wcica.2004.1340988

[2] Qing Hu,Yu-ju Zhan. Fault diagnosis technology based on SVM in power electronics circuit, Proceedings of the Chinese Society of Electrical Engineering, v 28, n 12, pp.107-111, 2008.

[3] Wang Zhengwu, Zhang Ruiping. Research on fault fuzzy diagnosis based on neural network. Journal of Systems Engineering, 2004, 34(11):88~93.

[4] Li Donghui, Zhou weiwei. Research on sensor fault diagnosis based on wavelet neural networks. Transactions of China Electortechnical Society, 2005, 20(5):49~52.

[5] Levin kieran T., Sun. Observer-based fault diagnosis of power electronics systems . 2010 IEEE Energy Conversion Congress and Exposition, ECCE 2010 - Proceedings, pp.4434-4440, 2010,.

DOI: 10.1109/ecce.2010.5618444