Fault diagnosis method for rolling bearings based on IZOA-VMD and Teager envelope spectrum

Abstract

A novel rolling bearing fault diagnosis approach integrating IZOA-VMD and the Teager envelope spectrum is proposed to overcome the inaccuracies of manual VMD parameter tuning. An Improved Zebra Optimization Algorithm (IZOA)— incorporating Sine chaotic mapping, elite opposition-based learning, and dynamic neighborhood local search—adaptively optimizes VMD. The Teager envelope spectrum is then applied to the decomposed modes to enhance fault signatures. Experimental validation on inner race, outer race, and rolling element defects demonstrates precise fault identification. This method effectively elevates decomposition accuracy and feature extraction, proving highly viable for practical engineering.