Variational mode decomposition applied to offshore wind turbine rolling bearing fault diagnosis

The incipient fault feature of offshore wind turbine rolling bearing is difficult to extract. A fault diagnosis method for offshore wind turbine rolling bearing based on variational mode decomposition(VMD) method is proposed. Considering the number of mode components has great influence on the decomposition results, the general principles of determination of the mode components number is proposed, which is based on analyzing the VMD decomposition of different modal components number impact on signal frequency characteristics. The number of mode components is determined by the proposed principles and got certain number of mode components . Then the best selected mode component is processed by envelop demodulation algorithm .The fault type of bearing is diagnosed by analyzing the envelop spectrum. The results of the standard rolling bearing fault data analysis shows that the proposed method can diagnose rolling bearing fault accurately. The incipient fault feature of offshore wind turbine vibration data is effectively extracted shows that the proposed method is significance of research and application.