Parameter optimized variational mode decomposition method with application to incipient fault diagnosis of rolling bearing

Abstract

The fault feature of rolling bearing in early failure period is difficult to extract. An incipient fault diagnosis method for rolling bearing based on parameter optimized variational mode decomposition method was proposed. Particle swarm optimization algorithm was used to search for the best combination of influencing parameters of variational mode decomposition algorithm, the penalty parameter and number of components were then set according to the searching results, and the fault signal was processed by parameter optimized variational mode decomposition algorithm. The original fault signal was decomposed into several intrinsic mode function components. The best signal component was selected and processed by envelope demodulation algorithm, the fault type of bearing was judged by analyzing the envelope spectrum of the signal. The simulated and measured signals of fault bearing were analyzed by parameter optimized variational mode decomposition method and the weak characteristic frequency information was extracted successfully. The results show that the proposed method enables to judge the incipient fault of rolling bearing effectively with desired reliability.