Diagnostics of Analog Circuits Based on LS-SVM Using Time-Domain Features

被引:56
作者
Long, Bing [1 ]
Li, Min [1 ]
Wang, Houjun [1 ]
Tian, Shulin [1 ]
机构
[1] Univ Elect Sci & Technol China, Sch Automat Engn, Chengdu 611731, Peoples R China
基金
中国国家自然科学基金;
关键词
Analog circuits; Diagnostics; Feature selection; Feature vector; Least squares support vector machine; Time-domain features; FAULT-DIAGNOSIS; WAVELET TRANSFORM; NEURAL-NETWORKS;
D O I
10.1007/s00034-013-9614-3
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Most researchers use wavelet transforms to extract features from a time-domain transient response from analog circuits to train classifiers such as neural networks (NNs) and support vector machines (SVMs) for analog circuit diagnostics. In this paper, we have proposed some new feature selection methods from a time-domain transient response, and compared the diagnostic results based on a least squares SVM (LS-SVM) using different time-domain feature vectors. First, we have improved two traditional feature selection methods: (a) using the mean and standard deviation in wavelet transform features, and (b) using the mean, standard deviation, skewness, kurtosis, and entropy in statistical property features. Then, a conventional time-domain feature vector based on the impulse response properties of a control system has been proposed. The simulation experiments for a leapfrog filter and a nonlinear rectifier show that: (1) the two improved methods have better accuracy than the traditional methods; (2) the proposed conventional time-domain feature vector is effective in the diagnostics of analog circuits-over 99 % for both of the two example circuits; (3) the proposed diagnostic method can diagnose soft faults, hard faults, and multi-faults, regardless of component tolerances and nonlinearity effects.
引用
收藏
页码:2683 / 2706
页数:24
相关论文
共 31 条
[1]   Analog fault diagnosis of actual circuits using neural networks [J].
Aminian, F ;
Aminian, M ;
Collins, HW .
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2002, 51 (03) :544-550
[2]   Fault diagnosis of nonlinear analog circuits using neural networks with wavelet and Fourier transforms as preprocessors [J].
Aminian, F ;
Aminian, M .
JOURNAL OF ELECTRONIC TESTING-THEORY AND APPLICATIONS, 2001, 17 (06) :471-481
[3]   Neural-network based analog-circuit fault diagnosis using wavelet transform as preprocessor [J].
Aminian, M ;
Aminian, F .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-ANALOG AND DIGITAL SIGNAL PROCESSING, 2000, 47 (02) :151-156
[4]   A modular fault-diagnostic system for analog electronic circuits using neural networks with wavelet transform as a preprocessor [J].
Aminian, Mehran ;
Aminian, Farzan .
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2007, 56 (05) :1546-1554
[5]  
Bing Long, 2008, 2008 International Conference on Communications, Circuits and Systems, P1026, DOI 10.1109/ICCCAS.2008.4657943
[6]   Machine Condition Prediction Based on Adaptive Neuro-Fuzzy and High-Order Particle Filtering [J].
Chen, Chaochao ;
Zhang, Bin ;
Vachtsevanos, George ;
Orchard, Marcos .
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2011, 58 (09) :4353-4364
[7]  
Chun-cheng Chen, 2010, 2010 IEEE 7th International Conference on Mobile Ad-Hoc and Sensor Systems (MASS 2010), P1, DOI 10.1109/MASS.2010.5663974
[8]   A novel approach of analog circuit fault diagnosis using support vector machines classifier [J].
Cui, Jiang ;
Wang, Youren .
MEASUREMENT, 2011, 44 (01) :281-289
[9]   Prognostics-based risk mitigation for telecom equipment under free air cooling conditions [J].
Dai, Jun ;
Das, Diganta ;
Pecht, Michael .
APPLIED ENERGY, 2012, 99 :423-429
[10]  
DeCarlo R., 1988, Circuits, Systems, and Signal Processing, V7, P481, DOI 10.1007/BF01599922