Wind generation set gearbox fault characteristic extraction method based on IVMD-SE

Abstract

The invention discloses a wind generation set gearbox fault characteristic extraction method based on IVMD-SE, and belongs to the technical field of the wind generation. The method comprises the following steps: respectively collecting original vibration signals of a wind generation set gearbox gear under a normal working status, an abrasion working status and a tooth breaking working status by utilizing a vibration acceleration sensor; optimizing a penalty term parameter a and a decomposition layer number K in the VMD parameter by adopting a PSO algorithm based on the integer planning; performing decomposition on the gear vibration signals under various working statuses by adopting an improved vibration modal method to obtain various modal components of the vibration signals; optimizingIMF components closely related to the original signals by utilizing the related coefficient method, and taking singular entropies of various modal components as wind turbine gearbox fault characteristics; inputting the characteristics into a multi-classification SVM to verify a characteristic extraction effect. The VMD and the singular entropies are combined to extract the wind generation set gearbox gear fault characteristics under the noise interference, the signal characteristics are strengthened, and the characteristic extraction effect is more obvious.

Images