Research On Preventive Maintenance Strategy Of Wind Turbine Based On Stochastic Differential Equation

Due to the maintenance development of science and technology,condition based maintenance(CBM)has been widely used in modern large-scale wind farm.Although the current condition based maintenance can maximize the availability of equipment through dynamic monitoring,in practice,many wind farms still mainly use time-based maintenance(TBM)and only use CBM for auxiliary maintenance.This makes the operation of wind turbines face problems such as over maintenance,under maintenance,frequent random faults and high maintenance cost.Based on this,this thesis describes the state change model of wind turbine based on stochastic differential equation(SDE),and deeply analyzes the temporal and spatial evolution relationship of wind turbine equipment.The main research contents are as follows:(1)This research introduces the monitoring data of the main components of the wind turbine,and uses the stochastic differential equation to describe the condition monitoring indicators of each equipment.This model divides the equipment state into two parts: internal degradation and external random interference.Polynomial approximation is used to construct the model of the equipment,and the external random interference is simulated by Brownian motion.At the same time,the parameter iteration method is used to solve the parameters of the equipment monitoring data model.(2)Based on the monitoring data of wind power equipment,the state evolution equation of wind power equipment deterioration is constructed.The downtime and maintenance rate of CBM are calculated by random analysis theory,the average maintenance rate of TBM is calculated by measure transformation,and the average value of track function of two maintenance strategies is calculated.Based on this,the general operation characteristics of TBM and CBM are analyzed.(3)Considering the introduction of the state evolution equation after maintenance,the optimal maintenance rate of the equipment is determined through the stochastic theory.Then,using the vector Markov theory,under the goal of the lowest maintenance cost,the periodic maintenance strategy and age maintenance strategy are adopted to find the optimal maintenance strategy and formulate the operation and maintenance strategy of the wind turbine.At the same time,through the characteristics of TBM and CBM,a TBM maintenance strategy based on CBM is proposed to ensure the advantages of TBM and CBM.The long-term space-time evolution characteristics of wind power equipment are analyzed.The research on the condition analysis and maintenance decision of wind turbine shows that TBM is a predictable and controllable process,which depends on the global data.CBM usually relies on local data,and its long-term behavior is unpredictable,which can not guarantee the long-term behavior of the system.Therefore,a single maintenance strategy is not optimal for wind turbines,but a combination of multiple strategies should be used,and periodic evaluation and feedback should be used to improve the maintenance strategy.The effectiveness of the proposed model and maintenance strategy is verified by an application example,which proves that the proposed strategy has a certain application value for the preventive maintenance of large-scale production process.