SVM Optimized by Immune Clonal Selection Algorithm for Fault Diagnostics

This paper presents a fault diagnosis method using Support Vector Machines (SVM) and Immune Clonal Selection Algorithm (ICSA). Support Vector Machines (SVM) has been well recognized as a powerful computational tool for nonlinear problems which have high dimensionalities. Whereas the parameters in SVM are usually selected by manpsilas experience, it has hampered the efficiency of SVM in practical application. Immunity Clonal Selection Algorithm (ICSA) is a new intelligent algorithm which can carry out the global search and the local search in many directions rather than one direction around the same individual simultaneously, and can effectively overcome the prematurity and slow convergence speed of traditional evolution algorithm. To improve the capability of the SVM classifier, we apply the immunity clonal selection algorithm to optimize the parameter of SVM in this paper. The experimental result shows that the fault diagnostics based on SVM optimized by ICSA can give higher recognition accuracy than the general SVM.