Research On Prediction Of Remaining Life Of Wind Turbine Based On Wiener Process

Wind energy,as the most potential energy for development,is valued by all countries.So the total installed capacity of wind turbines increases year by year.The wind turbine is composed of the transmission system,yaw system,paddle changing system and so on.The failure of any key component will lead to the shutdown of the wind turbine,which will seriously affect the annual operation time of the wind turbine.Because the wind farm is located in a remote and harsh environment,the wind turbine will be impacted by undirected,irregular and variable strength wind,resulting in frequent failures of the wind turbine.As a key component of wind turbine,the gearbox is one of the components that cause the most downtime.The gearbox is located at the top of the tower.When the gearbox causes the wind turbine to break down,the maintenance requires the help of large equipment,resulting in a high cost.Therefore,it is of great significance to conduct degradation modeling of the degradation state of the gearbox so as to predict its remaining life.It can ultimately provide relevant basis for the formulation of the maintenance strategy of the wind turbine.Due to the complex failure mechanism of the gearbox,it is difficult to accurately model its physical failure process by mathematical model.During the operation of the gearbox,a large amount of state data containing life information will be generated.Therefore,the data-based method is suitable for the degradation modeling and remaining life prediction of the gearboxIn this paper,based on the degradation state of the gearbox,a method based on Wiener process is used to model the degradation of the gearbox and predict the remaining life.The main research work and content are as follows:(1)Due to the repeated fluctuations of the gearbox degradation state in a certain range,the overall trend is an upward trend and nonlinear changes,so a degradation model based on a nonlinear Wiener process is used for modeling.At the same time,the measurement error in the process of degraded data acquisition is considered,and a real-time degraded state update model considering the measurement error is established;the unknown parameters in the model are estimated using the method of maximum likelihood estimation;the state space model is established,and the method of Kalman filtering,Real-time update prediction of its degradation state;According to the concept of the first hitting time,the remaining life probability density function is obtained,thereby obtaining the remaining life,and finally the gear fatigue test data is used to verify the above method and prove its effectiveness.(2)Consider the influence of wind speed on the degradation state of the gearbox,express the influence of the external environmental wind speed on the degradation of the gearbox as a power law model,and establish the degradation model of the gearbox;adopt the two-step method of maximum likelihood estimation and least square method Estimate the unknown parameters in the model;use the concept of first hitting time to derive the expression of the remaining life probability density function and get the remaining life;finally,the effectiveness of the method is verified by simulation.