Fault Diagnosis of Wind Turbine Planetary Gearbox Based on Variational Mode Decomposition and Convolution Depth Belief Networks

The vibration signal of planetary gearbox of wind turbine is a kind of non-linear and non-stationary complex signal. The traditional fault diagnosis method can deal with this kind of signal well in a limited range. In this paper, an intelligent diagnosis method based on the combination of variable mode decomposition (VMD) and convolutional deep belief networks (DCBN) is proposed, and the field of gearbox fault detection is introduced. Firstly, VMD is used to decompose the original signal, and the eigenmode function with larger impact content is extracted by kurtosis criterion. Then the obvious components of the feature information are fused together to form the multi-channel input, and then dcbn is used to extract the feature. Finally, the feature is input into the classifier for fault identification and classification. The experimental results show that the method can accurately classify the planet of wind turbine Different working conditions and fault types of gearbox are identified and classified.