Research On State Modeling And Remaining Useful Life Prediction Of Wind Turbine Gear Box

Wind energy is one of the most important energy sources,and its installed capacity is increasing rapidly in recent years.Wind turbine is a typical series repairable system composed of transmission system,yaw system and variable pitch system.Wind turbine is often affected by irregular wind impact,temperature and weather changes,etc.,which leads to frequent failures of wind turbine.The gearbox system is the key component of wind turbine.Its function is to transfer the power generated by the wind vane to the generator and to make the speed of the generator reach the corresponding value.At the same time,it is also one of the most frequently faulty parts in the wind turbine.The gearbox is usually installed at the top of the tower.When the gearbox fails,the gearbox needs to be disassembled and repaired,which results in high maintenance cost.If the maintenance is not timely,it will accelerate the aging or even malfunction of the bearing on both sides of the gearbox,which will lead to the serious accident of the wind turbine.Therefore,it is of great significance to model the running state of wind turbine gearbox and predict its remaining useful life.In order to solve this problem,this paper first analyzes the historical fault data of the gearbox of wind turbine,and establishes a model of the development trend of the gearbox running state and performance degradation by using the condition monitoring data of a long time before the failure.The gearbox system can be divided into three stages according to the running state: the fault incubation period,the fault development period and the fault period.The fault development period reflects the degradation speed and performance degradation trend of the equipment.Because the state monitoring quantity which reflects the fault characteristics fluctuates repeatedly in a certain range,a performance degradation model based on Wiener process is used to simulate the degradation process of the gearbox.Firstly,the trend quantity which can reflect the degradation of equipment performance is constructed by using the monitoring quantity of temperature and rotational speed state in different monitoring periods of gearbox,which can meet the requirement of Wiener distribution,and its reliability is tested by probability distribution.Secondly,the performance degradation model of gearbox based on Wiener process is obtained by maximum likelihood function estimation.On this basis,the residual service life prediction model of wind turbine gearbox is established from the point of view of probability and statistics,and the remaining service life of the gearbox is obtained by taking the first time of reaching the failure threshold in the degradation process as the gearbox failure time.Finally,based on the monitoring data of the maximum wind month and the minimum wind month in the area where the wind farm is located,the prediction interval of the remaining service life of the gearbox is obtained by modeling the degradation process of the gearbox,respectively.The reliability of the model is calculated.Finally,using MATLAB platform,a wind turbine gearbox state evaluation model and residual service life prediction model are established.The gearbox data in different stages are tested,and the simulation results are compared with the actual life.The effectiveness of this method is verified.