Kernel principal component analysis fault diagnosis method based on sound signal processing and its application in hydraulic pump

In situations where the vibration sensor is not suitable to be used and while fault diagnosis method based on vibration signal processing has limitations, KPCA fault diagnosis method based on sound signal is proposed. The basic theory of kernel principal component analysis and its basic procedures for fault detection are introduced and sound signal pre-processing is depicted, multi-domain feature vector is extracted from time, time-frequency and frequency domain, faults are diagnosed with kernel principal component analysis method. The new kernel principal component analysis fault diagnosis method based on sound signal processing is tested on axial piston pump, its result shows that the method is effective and it can overcome deficiencies of the fault diagnosis method based on vibration signal.