A NEW ROLLER BEARING FAULT DIAGNOSIS METHOD BASED ON MVMD-RMS AND DE-LSSVM

Tóm tắt

This research introduces a novel method for roller bearing fault diagnosis based on the Least Squares Support Vector Machine (LSSVM) with parameters optimized using the Differential Evolution (DE) algorithm, referred to as DE-LSSVM. Initially, the Multivariate Variational Mode Decomposition (MVMD) method decomposes the acceleration vibration signals from roller bearings into component functions. Subsequently, these functions extract initial feature matrices using the root mean square (RMS) method. Finally, these values serve as input vectors for the DE-LSSVM classifier. Experimental results illustrate that the proposed method exhibits lower test error rates and reduced computational time when compared to other methods using the same collected data.

DOIs: https://doi.org/10.46242/jstiuh.v70i4.4763