Fault Diagnosis of Analog Circuit Based on Multi-Test Points and Multi-Feature Information

Abstract:

In the process of use BP neural network to fault diagnosis of analog circuits, the network input which represents fault signature was very important. A given new method which base on multi-points and multi-feature information is taken to construct the original sample set. With this method to construct the original fault signature set, then as the input of BP neural network and train the network. Simulation results show that, the network train with sample set which constructed by this method use in fault diagnosis of analog circuits is better accuracy than traditional methods. Proved the feasibility of this method in fault diagnosis of analog circuits, and offer a new method for fault diagnosis of analog circuits.

• Periodical: Applied Mechanics and Materials (Volumes 313-314)
• Pages: 277-280
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.313-314.277
• Online since: March 2013
• Authors: Qiang Pan, Chao Yang
• Keywords: Analog Circuits, BP Neural Network (BPNN), Fault Diagnosis, Fault Feature

References

[1] Wang Weidong. Principle of modern analog integrated circuits and application [M] Beijing: Electronic Industry Press, 2008.

[2] Qu Yi Ning Duo, Huang Jianbing, based on the value and the BP algorithm analog circuit fault diagnosis [J]. Computer Measurement & Control, 2009,17 (12) :2417-2419.

[3] Chen Min, Zhang Buhan, Duan Xianzhong, based on the minimum singular value of the sensitivity of the voltage stability weak node study [J] Power System Technology, 2006,30 (24) : 36-39.

[4] EBerhart R C, Dobbins R W. Neural Network PC Tools: A Practical Guide [M]. Academic Press, 1990.

[5] Kim Yu analog circuit fault diagnosis method based on wavelet neural network research [D]. Chengdu: University of Electronic Science and Technology in 2008.