Research on Diesel Engine Bearing Fault Diagnosis based on Vibration Signal

The vibration signal of diesel engine bearing has the characteristics of nonlinear and is easily affected by the coupling of noise and vibration. In this paper, we use variational modal decomposition and the Gray Wolf algorithm to optimize support vector machines for fault diagnosis. Firstly, VMD is used to decompose and preprocess vibration signals, and the intrinsic modal functions (IMF) of different scales are obtained. Then, the energy value and energy entropy of decomposed signals are extracted to construct feature vectors. Finally, SVM and GWO optimized SVM are used for fault identification and classification respectively. In addition, to deal with the problem that VMD parameters are difficult to choose correctly, this paper adopts SSA algorithm to optimize VMD decomposition parameters and constructs AVMD-GMO-SVM model. The experimental data of bearings show that the proposed fault diagnosis model has high fault pattern recognition accuracy and can effectively realize the recognition and classification of diesel bearings.