Fault Diagnosis Algorithm Based on Artificial Immunity System

Artificial immunity systems have been emerged as simulation of human immunity. The negative selection algorithm which is most important component of artificial immunity system can determine undesired condition, easily. In this study, motor current signature analysis and negative selection algorithm have been used for broken rotor bar faults. Current signal obtained from motor has been transformed to current spectrum by using motor current signature analysis. The side bands extracted from this spectrum have been taken as input to negative selection algorithm, and broken rotor bar faults have been diagnosed. The application of developed fault diagnosis algorithm has been demonstrated by diagnosing faults in induction motor real time. Furthermore, proposed fault diagnosis approach is adapted for diagnosing other faults

[1]  Mo-Yuen Chow,et al.  Neural networks-based negative selection algorithm with applications in fault diagnosis , 2004, 2004 IEEE International Conference on Systems, Man and Cybernetics (IEEE Cat. No.04CH37583).

[2]  H. Joel Trussell,et al.  Fuzzy inference systems implemented on neural architectures for motor fault detection and diagnosis , 1999, IEEE Trans. Ind. Electron..

[3]  Mo-Yuen Chow.  Methodologies of Using Neural Network and Fuzzy Logic Technologies for Motor Incipient Fault Detection , 1998 .

[4]  João F. Martins,et al.  Fault detection using immune-based systems and formal language algorithms , 2000, Proceedings of the 39th IEEE Conference on Decision and Control (Cat. No.00CH37187).

[5]  B. Ayhan,et al.  Multiple signature processing-based fault detection schemes for broken rotor bar in induction motors , 2005, IEEE Transactions on Energy Conversion.

[6]  Xiao Zhi Gao,et al.  Soft computing methods in motor fault diagnosis , 2001, Appl. Soft Comput..