Abstract:It is known that the wavelet-singular point detection-based method is sensitive to noises; to solve this problem, a method of fault detection for bearings based on wavelet transform modulus maximum Lipschitz spectrum entropy is proposed by combining wavelet analysis with entropy theory, including the detection scheme of bearing vibration faults and the threshold selection method based on swarm intelligence. The proposed method is compared with the methods based on wavelet energy spectrum and wavelet packet energy spectrum entropy and the wavelet-singular point detection-based method in the experiments. The results show that the proposed method is particularly well adapted to describe fault characteristics and fault diagnosis, which outperforms the other three methods in terms of detection time and detection rate.