Fault Diagnosis of Motor Bearing Based on Improved SSA Optimized VMD Combined with Convolution Neural Network

In order to solve problems that swarm intelligence optimization algorithm is easy to fall into local optimization when optimizing the parameters of variational mode decomposition, which leads to incomplete feature extraction, which leads to low accuracy of subsequent fault diagnosis, a motor bearing fault diagnosis method based on improved sparrow algorithm optimization variational mode decomposition combined with bidirectional long short-term memory-convolution neural network was proposed. Firstly, the sparrow population was initialized by Logistic chaotic map to make the population randomly and uniformly distributed in the search space. Secondly, the global exploration strategy of Osprey algorithm was introduced to replace the explorer position update formula of Sparrow algorithm. Finally, the Cauchy mutation strategy was integrated into the follower position update of Sparrow algorithm, which improves the global search ability of the algorithm and avoids falling into local optimization. The algorithm was applied to the parameter optimization of variational mode decomposition, and the method of convolution neural network was used to classify the bearing fault signals. the experimental results show that this method can achieve fast and accurate fault diagnosis.