Fault Diagnosis of Wind Turbine Gearbox Based on SSDAE-ELM

In recent years, the installed capacity of wind turbine is increasing year by year, and the influence of wind turbine faults on wind farm operation is also increasing. In this paper, a novel fault diagnosis method is proposed to diagnose the fault of wind turbine gearbox. The method is based on stacked sparse denoising auto-encoders (SSDAE) neural network which combines stacked sparse auto-encoders (SSAE) neural network and stacked denoising auto-encoders (SDAE) neural network to extract features from high-dimensional signal data better. In addition, extreme learning machine (ELM) classifier is used to finish the fault diagnosis according to the features extracted from high-dimensional signal data by the SSDAE. Finally, the method is tested based on the simulation experiment, the simulation results demonstrate that the accuracy of this fault diagnosis method is higher than that of other methods when the same parameters are used.