Aiming at the complex vibration signals of rolling bearings and the difficulty in fault diagnosis, a fault diagnosis method based on the EEMD-MDE-improved binary tree SVM hybrid algorithm is proposed. First, the ensemble empirical mode decomposition is used to decompose the original signal into multiple simple component signals. Secondly, the feature vector is constructed using multi-scale spread entropy. Then, a calculation method to deal with unbalanced sample data is proposed, and the method is combined with SVM, and an improved binary tree SVM algorithm is proposed. The analysis and verification of the measured signals of the bearings of Western Reserve University show that the average accuracy of the method in this paper is as high as 98% in the fault diagnosis of rolling bearings, which proves the superiority of the method in this paper.
Fault Diagnosis of Bearing Based on EEMD-MDE-Improved Binary Tree SVM Hybrid Algorithm
2023-10-11
2582819 byte
Conference paper
Electronic Resource
English
Urban Rail Train Wheel Fault Diagnosis Based on Improved EEMD
British Library Conference Proceedings | 2020
|Urban Rail Train Wheel Fault Diagnosis Based on Improved EEMD
Springer Verlag | 2020
|Urban Rail Train Wheel Fault Diagnosis Based on Improved EEMD
TIBKAT | 2020
|A Fault Diagnosis Method of Gear Based on SVD and Improved EEMD
Springer Verlag | 2017
|