Optimised CEEMDAN based on sparrow search algorithm for Motor Bearing Fault Feature Extraction

For fault diagnosis, it is very important to extract effective fault feature information.Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) is widely used in fault diagnosis due to its advantages of fast decomposition speed, good completeness of results, and high adaptivity. However, when the added Gaussian white noise is not suitable, it will still be affected by problems such as modal aliasing and component signal distortion, which increase the difficulty of diagnosis. In this paper, the Sparrow Search Algorithm (SSA) is used to find out the optimal Gaussian white noise amplitude and the number of additions required for different fault information, so as to optimise the processing capability of CEEMDAN and achieve more accurate fault diagnosis. The experimental results show that this method suppresses the interference of modal aliasing and spurious components to a certain extent, and enhances the effectiveness of bearing fault feature extraction.