Fault isolation in analog circuits using a fuzzy inference system

Abstract This paper introduces a fuzzy inference system (FIS) for single analog fault diagnosis. The ability of fuzzy logic to encode structured knowledge in a numerical framework is exploited in isolating faults in analog circuits. A training set that simulates the behaviour of the circuit due to a set of anticipated single faults as well as the fault-free situation is first constructed. For each anticipated fault, this set relates the circuit measurements to the corresponding deviation in the faulty circuit element from its nominal. These measurements and the deviations in circuit elements are both fuzzified into appropriate linguistic fuzzy values. A fuzzy rule base for each fault that characterizes the circuit response by linking symptoms to causes is built. The outputs of the fuzzy rule bases are then defuzzified to recover crisp values for the deviations in circuit elements. A fault diagnosis procedure that utilizes the proposed FIS is also presented along with a brief analysis and comparison with a number of existing artificial intelligence-based techniques. A test example that demonstrates the potential of this procedure in fault isolation is illustrated.

[1]  Alessandro Giua,et al.  Qualitative dynamic diagnosis of circuits , 1993, Artif. Intell. Eng. Des. Anal. Manuf..

[2]  Benjamin J. Kaipers,et al.  Qualitative Simulation , 1989, Artif. Intell..

[3]  Meryem Marzouki,et al.  Test and diagnosis of analog circuits: When fuzziness can lead to accuracy , 1996, J. Electron. Test..

[4]  Mohamed A. El-Gamal,et al.  A knowledge-based approach for fault detection and isolation in analog circuits , 1997, Proceedings of International Conference on Neural Networks (ICNN'97).

[5]  Hany L. Abdel-Malek,et al.  A new approach for the selection of test points for fault diagnosis , 1995, Proceedings of ISCAS'95 - International Symposium on Circuits and Systems.

[6]  J.W. Bandler,et al.  Fault diagnosis of analog circuits , 1985, Proceedings of the IEEE.

[7]  J.A. Starzyk,et al.  Artificial neural network for testing analog circuits , 1990, IEEE International Symposium on Circuits and Systems.

[8]  Mohamed Fathy Abu El-Yazeed,et al.  A Combined Clustering and Neural Network Approach for Analog Multiple Hard Fault Classification , 1999, J. Electron. Test..

[9]  LiMin Fu,et al.  Knowledge base refinement by backpropagation , 1991, Data Knowl. Eng..

[10]  Mongi A. Abidi,et al.  Fuzzy logic-based data integration: theory and applications , 1994, Proceedings of 1994 IEEE International Conference on MFI '94. Multisensor Fusion and Integration for Intelligent Systems.

[11]  Corinne Bos-Plachez,et al.  A possibilistic ATMS contribution to diagnose analog electronic circuits , 1997, International Journal of Intelligent Systems.

[12]  L. Zadeh,et al.  Fuzzy sets and applications : selected papers , 1987 .

[13]  C. R. Parten,et al.  Fault diagnosis and neural networks , 1991, Conference Proceedings 1991 IEEE International Conference on Systems, Man, and Cybernetics.

[14]  Antonio Torralba,et al.  Fault detection and classification of analog circuits by means of fuzzy logic-based techniques , 1995, Proceedings of ISCAS'95 - International Symposium on Circuits and Systems.

[15]  I. M. Bell,et al.  Neural network approach to fault diagnosis in CMOS opamps with gate oxide short faults , 1994 .

[16]  Mohamed El-Gamal.  Fault location and parameter identification in analog circuits , 1990 .

[17]  M. Abdulghafour,et al.  A fuzzy logic system for analog fault diagnosis , 1996, 1996 IEEE International Symposium on Circuits and Systems. Circuits and Systems Connecting the World. ISCAS 96.

[18]  Peter Y. K. Cheung,et al.  Analog fault diagnosis-A practical approach , 1994, Proceedings of IEEE International Symposium on Circuits and Systems - ISCAS '94.

[19]  Benjamin Kuipers,et al.  Qualitative Simulation , 1986, Artificial Intelligence.

[20]  中園 薫.  A Qualitative Physics Based on Confluences , 1986 .

[21]  B. Kaminska,et al.  Fault observability analysis of analog circuits in frequency domain , 1996 .

[22]  G. Rutkowski.  A neural network approach to fault location in nonlinear dc circuits , 1992 .

[23]  S. Bedrosian,et al.  Fault isolation algorithm for analog electronic systems using the fuzzy concept , 1979 .

[24]  Paul W. Tuinenga,et al.  SPICE: A Guide to Circuit Simulation and Analysis Using PSpice , 1988 .

[25]  Anil Pahwa,et al.  Band faults: Efficient approximations to fault bands for the simulation before fault diagnosis of linear circuits , 1982 .

[26]  Corinne Bos-Plachez.  A possibilistic ATMS contribution to diagnose analog electronic circuits , 1997 .

[27]  Shigeo Abe,et al.  Neural Networks and Fuzzy Systems , 1996, Springer US.

[28]  Benjamin R. Epstein,et al.  Fault detection and classification in linear integrated circuits: an application of discrimination analysis and hypothesis testing , 1993, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[29]  Mongi A. Abidi,et al.  Fusion of visual and range features using fuzzy logic , 1994 .