Research On Motor Bearing Fault Diagnosis Based On Improved Random Forest

Motor bearings are the most common transmission parts of various mechanical and electrical equipment.They are widely used in pillar industries such as mechanical manufacturing and processing,coal mining and washing,electric power and thermal production,etc.To ensure the normal operation of motor bearings can guarantee the production safety of enterprises and improve the production efficiency of enterprises.Based on vibration signals,this paper studies fault feature extraction and fault feature pattern recognition of motor bearings,and the research contents are as follows:(1)Relevant basic theories.On the basis of explaining the structure and fault mechanism of the motor bearing,the common fault diagnosis methods,fault signal feature extraction methods and fault pattern recognition methods of the motor bearing were analyzed and studied,which laid a theoretical foundation for the subsequent research.(2)Motor bearing fault signal feature extraction.In order to solve the problem of signal distortion and mode aliasing in the aggregated empirical mode decomposition method,the sparrow search algorithm is proposed to optimize the decomposed noise components to complete signal reconstruction.The experimental results show that the improved empirical mode decomposition method can extract the effective signal features.(3)Establishment of motor bearing fault diagnosis model.In view of the overfitting of traditional random forest algorithm in complex cases,the sparrow search algorithm was used to optimize the number of trees and the minimum number of leaf nodes in the random forest,and the traditional random forest algorithm was improved,and a fault diagnosis model based on the improved random forest algorithm was established.(4)Motor bearing fault diagnosis.Firstly,an improved aggregated empirical mode decomposition method is used to extract signal fault features and construct feature vectors to form a sample set.Then,the sample set is input into the fault diagnosis model based on the improved random forest algorithm,and the fault discrimination is carried out.The results show that compared with the traditional random forest algorithm model and the two optimized neural network algorithm models,the improved random forest algorithm diagnosis model has improved the diagnosis accuracy for various fault types,which verifies the effectiveness of the proposed scheme.