Research on fault diagnosis of rolling bearing based on wavelet packet energy feature and planar cloud model

Due to the uncertainties such as the strong randomness of feature information in the process of the rolling bearing fault diagnosis, misjudgment usually exists, and it causes reduced accuracy of the diagnosis. Thus the paper proposes a wavelet packet energy feature (WPEF) and planar cloud model (PCM) fault diagnosis method, PCM has the advantages of stable tendency and the ability in order to organically combine the fuzziness and the randomness. Firstly, the cloud models of fault concepts are obtained by using the wavelet packet to extract the energy characteristics of training sample from different fault. Then, the tested fault sample is used to extract the energy characteristics of wavelet packet, it is regarded as cloud droplets, then to classify it to realize fault diagnosis. The comparative experiments show the effectiveness and feasibility of the proposed method.