Fault Diagnosis of Rolling Bearing of Drilling Rig Based on Optimized VMD and CNN-BiLSTM

Abstract

Rolling bearings are key basic components of oil electric drilling rig equipment, and diagnosing faults in these components is of great significance to ensure the safe operation of the equipment. This paper presents a scheme to identify the fault state of rolling bearings based on the combination of variational mode decomposition and the CNN-BiLSTM neural network, using the Sparrow Search Algorithm. The multi-strategy improved Sparrow Search Algorithm is employed to solve the problem of determining VMD parameters. Then, the CNN-BiLSTM neural network model is introduced for fault diagnosis classification. By comparing the accuracy of four fault diagnosis models, it is demonstrated that the improved Sparrow Search Algorithm based on multi-strategy is used to optimize VMD. When combined with the CNN-BiLSTM network model, the accuracy of fault diagnosis can reach 98.3%, demonstrating the ability to accurately identify fault types.