Enhanced symplectic characteristics mode decomposition method and its application in fault diagnosis of rolling bearing

As an adaptive signal decomposition method, symplectic geometry mode decomposition (SGMD) method is suitable for dealing with non-stationary signals However, the decomposition effect is not ideal when dealing with rolling bearing fault signals with strong background noise. On the one hand, this noise reduction method of SGMD is not suitable for fault signals with strong background noise. On the other hand, SGMD uses QR decomposition method, which results in decomposition error diffusion in the decomposition of singular matrix. Therefore, an enhanced symplectic characteristics mode decomposition (ESCMD) method is proposed in this paper. ESCMD enhances fault features through the calculus operator to make fault features easier to extract, and replaces QR decomposition with eigenvalue decomposition (EVD) to avoid error diffusion during matrix decomposition. Emulational and experimental results show that ESCMD has excellent noise robustness and feature enhancement performance. (C) 2020 Published by Elsevier Ltd.