Research On Vibration Characteristics And Fault Diagnosis Method Of Wind Turbine Transmission System

Gear transmission system is one of the important components of wind turbines,the operating environment of wind turbines is usually harsh,The sudden change of wind speed,dust,water vapor erosion and other factors often lead to gear cracks,gear peeling and other faults in the gear transmission system,Affect the normal operation of wind turbine equipment,cause serious economic losses.Therefore,early weak faults can be detected in time,and maintenance can be carried out through appropriate planned shutdown,which can reduce the maintenance cost of the fan and increase economic benefits.Collecting gear vibration signals for fault diagnosis is a widely used method in the field of gear fault diagnosis,and many methods have been developed for fault diagnosis of gear transmission systems.In addition,through the dynamic modeling of the gear transmission system,the simulation analysis of the vibration characteristics of the faulty gear is also helpful for the study of the early weak fault of the gear.In this paper,the wind power gear transmission system is taken as the research object,the dynamic modeling of the gear transmission system is carried out,and the main excitation-stiffness excitation on the gear transmission system is studied.By taking into account the dynamic model of stiffness excitation,the vibration characteristics of the gear transmission system under normal and fault conditions are analyzed,and the model is verified according to the experimental bench data.Based on the feature fusion fault diagnosis method of convolutional neural network,the vibration fault of the gearbox transmission system test bench is diagnosed.This paper mainly carries out the following work:(1)Taking the one-stage planetary gear train and the two-stage fixed-axis gear train of the wind power gear transmission system as the research objects,the translation-torsion coupling dynamic model of the gear transmission system is established,and the dynamic differential equation of the system is deduced.And the structural characteristics of the system are analyzed.(2)The energy method is used to calculate the time-varying mesh stiffness of gears,and the situation that the root circle and base circle of gears are not equal is discussed.On this basis,the gear crack fault and gear spalling fault are studied,which lays a foundation for the subsequent dynamic analysis.(3)Substitute the stiffness excitation into the dynamic model of the gear transmission system,simulate and analyze the vibration response of the gear under normal and fault conditions,and conduct the corresponding frequency domain analysis.The model is verified by using the experimental signals of the experimental bench.(4)The feature fusion fault diagnosis method based on neural network is adopted,and the sparse resonance decomposition algorithm and the Teager energy operator are used to extract and highlight the periodic component and shock component of the vibration data,which are used for the fault diagnosis of the convolutional neural network.