Skip to Main Content
For blindness of the parameter settings in kernel principal component analysis (KPCA), kernel function parameter optimized by particle swarm optimization algorithm (PSO) is proposed, and KPCA is applied to feature extraction. The mathematical model of kernel function parameter optimized is constructed firstly, then the particle swarm optimization algorithm with adaptive accelerate (CPSO) is used to optimize it. The optimized KPCA is applied to feature extraction of gearbox typical faults. The results indicate that KPCA after parameter optimized can effectively reduce the dimensions of feature vector of gearbox, and it has a better fault classification performance than linear principal component analysis (PCA). This method has an advantage in nonlinear feature extraction of mechanical failure signal.