Research on Wind Turbine Gearbox Fault Diagnosis Based on CEEMDAN and CVFDT

Gearbox fault diagnosis has always been a hot topic in the field of mechanical fault diagnosis and has great research value and significance. But at present, it is mainly based on offline data for fault diagnosis, which is not real-time and slow to find. In this paper, the real-time diagnosis of gearbox bearing fault is realized based on flow data. Aiming at the problems of nonstationary and weak fault characteristics of gearbox bearing vibration signals collected in real time, an energy entropy feature extraction method based on sliding window is proposed for adaptive noise complete set empirical mode decomposition (SW-CEEMDAN), which not only effectively solves the problem of endpoint effect of CEEMDAN, The feature vectors containing rich fault features can also be extracted; Then, the incremental training and adaptive update of the model are carried out by combining the incremental classification method CVFDT, and then the gearbox fault diagnosis can be realized in real time. Finally, the validity of the improved method was proved by the bearing data set of Case Western reserve University.