Fault Condition Recognition Based on PSO and KPCA

摘要

A method of kernel principal component analysis (KPCA) based on particle swarm optimization (PSO) is presented, which is applied in fault condition recognition of gear box. Comprehensively considered within-class scatter and between-class scatter of samples feature, the fitness function of kernel function parameter optimized is constructed, and the particle swarm optimization algorithm with adaptive accelerate (CPSO) is applied to optimize it. This method is applied to gear box condition recognition, compared with the recognized results based on principal component analysis (PCA). The results show that KPCA optimized by CPSO can effectively recognize fault conditions of gear box by reducing bind set-up of kernel function parameter, and its results of fault recognition outperform those of PCA. The conclusion is that KPCA based on PSO has advantage in nonlinear feature extraction of mechanical failure, and is helpful for fault conditional recognition of the complicated machine.