Analog circuits fault diagnosis using energy information of wavelet packet coefficients

被引：0
作者：
Luo, Hongping [1 ]
Li, Penghua [1 ]
Luo, Dechao [2 ]
Li, Yuanyuan [1 ]
机构：
[1] Automotive Electronics Engineering Research Center, College of Automation, Chongqing University of Posts and Telecommunications, Chongqing
[2] Key Laboratory of Vehicle Emission & Economizing Energy, National Institute of Automotive Engineering, Chongqing
来源：
Journal of Computational Information Systems | 2015年 / 11卷 / 08期
基金：
中国国家自然科学基金;
关键词：
Analog circuits; Energy calculation; Fault diagnosis; Wavelet packet transform;
D O I：
10.12733/jcis13780
中图分类号：
学科分类号：
摘要：
This paper presents a systematic approach to automate the fault diagnostic process of the actual circuits. The original signals are extracted, using a data acquisition board, from the output terminals of the circuits under test. The collected data is preprocessed by wavelet packet decomposition. A new energy function is defined to compute the energy values of the wavelet coefficients for generating the fault features. These energy features are translated into an appropriate form via a sensible encoding rule, then fed to the discrete Hopfield neural network (DHNN). The feature codes extracted from actual circuits and the SPICE simulations are used as the initial states and the attractors of the DHNN, respectively. The network is subject to some iterations using synchronous updating which is stopped after a while. The fired neurons read out to see which fault pattern is in this network. The results demonstrate that our fault diagnostic system has an acceptable high accuracy for the soft faults and hard faults. ©, 2015, Binary Information Press. All right reserved.
引用
收藏
页码：2795 / 2803
页数：8
相关论文
共 14 条
  • [1] Yuan L., He Y., Huang J., Sun Y., A new neural-network-nased fault diagnosis approach for analog circuits by using kurtosis and entropy as a preprocessor, IEEE Trans. Instrum. Meas., 59, 3, pp. 586-595, (2010)
  • [2] Aminian M., Aminian F., A modular fault-diagnostic system for analog electronic circuits using neural networks with wavelet transform as a preprocessor, IEEE Trans. Instrum. Meas., 56, 5, pp. 1546-1554, (2007)
  • [3] Spina R., Upadhyaya S., Linear circuit fault diagnosis using neuromorphic analyzer, IEEE Trans. Circuits Syst. II, 44, 3, pp. 188-196, (1997)
  • [4] Aminian M., Aminian F., Neural-network based analog-circuit fault diagnosis using wavelet transform as preprocessor, IEEE Trans. Circuits Syst. II, 47, 2, pp. 151-156, (2000)
  • [5] Aminian F., Aminian M., Collins H.W.J., Analog fault diagnosis of actual circuits using neural networks, IEEE Trans. Instrum. Meas., 51, 3, pp. 544-550, (2002)
  • [6] Xiao Y., He Y., A novel approach for analog fault diagnosis based on neural networks and improved kernel PCA, Neurocomputing, 74, pp. 1102-1115, (2011)
  • [7] Xiao Y., Feng L., A novel linear ridgelet network approach for analog fault diagnosis using wavelet-based fractal analysis and kernel PCA as preprocessors, Measurement, 45, pp. 297-310, (2012)
  • [8] Hopfield J.J., Neural networks and physical systems with emergent collective computational abilities, Proc. Natl. Acad. Sci. USA, 79, 8, pp. 2554-2558, (1982)
  • [9] Wen U., Lan K., Shih H., A review of Hopfield neural networks for solving mathematical programming problems, European Journal of Operational Research, 198, pp. 675-687, (2009)
  • [10] Du K.L., Swamy M.N.S., Neural Networks in a Softcomputing Framework, (2006)
12