A fault diagnosis approach for rolling bearing based on Fourier transform multi-filter decomposition and symbolic dynamic entropy

Bearings are one of the important components of mechanical systems, and their state affects the system operating performance, production efficiency and manufacturing quality, therefore, it is necessary for bearing fault prediction and diagnosis. In this paper, a fault diagnosis approach based on Fourier transform multi-filter decomposition (FTMFD) and symbolic dynamic entropy (SDE) is proposed. To obtain more fault information, FTMFD is proposed for signal decomposition. This decomposition method is then combined with SDE to obtain the fault features. Simulation of the proposed decomposition method is presented in the paper. The bearing dataset from Case Western Reserve University bearing data center is used to verify the effectiveness of the proposed diagnostic approach with three different classifiers. The experimental results show that the proposed approach has excellent fault diagnosis capability.