Research On Fault Diagnosis Of Wind Turbine Gearbox

Wind turbine works in the severe natural environment,making it easy to generate faults.Among them,the gearbox has become a frequent component of failure due to its own structural characteristics and operating conditions.At the same time,once the wind turbine occurred fault,it will cause great economic loss and severe social impact.Aimed above condition,wind turbine gearbox was selected to carry out research on vibration monitoring and fault diagnosis of wind turbine gearbox.The main research contents are as follows:1.The main structure of wind turbine gearbox is analyzed in this paper.Aimed at wind turbine gearbox structure,the dynamic modeling of wind turbine gearbox is established and analyzed,and then dynamic characteristics of wind turbine gearbox are understood deeply,which lay a theoretical basis for the following of damping noise reduction.2.The formation causes and failure mechanism of fault modes of wind turbine gearbox are analyzed.Aimed at fault modes of wind turbine gearbox,emulation signals of fault modes of wind turbine gearbox are established.Logical contact between failure characteristics and fault modes are built via researching failure mechanism and vibration characteristics of wind turbine gearbox,which offer a great help for selecting feature extraction method,establishing diagnosis knowledge base and designing gearbox fault diagnosis system.3.According to the above research of vibration signal of wind turbine gearbox,a fault diagnosis method based on the kernel-domain spectrum is proposed in this paper.In order to verify the effectiveness of kernel-domain spectrum method,analog signal and measured signal of wind turbine gearbox are used to verify it respectively.And then the feasibility of kernel-domain spectrum fault diagnosis method for small sample size is explored via analyzing experiment results.What's more,this method is improved on object template and a complete set of kernel-domain spectrum fault diagnosis method is established.4.A fault diagnosis method based on integral extension load mean decomposition multiscale entropy and least squares support vector machine is proposed in this paper.Thismethod solves the problem of low signal-to-noise ratio and difficult identification of wind turbine gearbox effectively.The validity of the method was verified by the measured signal,which provided a powerful help for the fault identification of the wind turbine gearbox.