| With the rapid growth of wind power installed capacity,the wind powers operation and maintenance industry has also developed rapidly.Wind farms are characterized by remote locations and complex and changeable external influences.This makes the operation of wind turbines faces frequent failures,low reliability and high maintenance costs.Although the current condition based maintenance(CBM)reduces accidental failures through dynamic maintenance and avoids the over maintenance and insufficient maintenance caused by the unreasonable time of the traditional time based maintenance(TBM),the traditional CBM’s monitoring system also requires high investment,and the internal relationship between CBM and TBM maintenance time needs to be analyzed.In response to the above problems,this paper proposes a stochastic degradation model(Stochastic Differential Equation,SDE)to describe the dynamic degradation process of wind turbines,and then analyzes CBM optimization methods in depth based on SDE,and achieves the purpose of reducing maintenance costs and improving the reliability of wind turbines by analyzing CBM optimization methods.The main research contents are as follows:(1)Establish a wind turbine degradation model based on stochastic differential equations.First,the existing reliability model is analyzed,combined with the degradation characteristics of wind turbines,and SDE is proposed to represent the change process of the wind turbine status.The model uses ordinary differential equations to describe the natural degradation process,and Brownian motion describes the impact of external disturbances on the performance of wind turbines.Then solve the SDE according to the Ito principle to obtain the wind turbine state expression,and finally obtain the model parameters according to the nature of the SDE solution.(2)For the single-component system,the optimization method of CBM is studied.First,the overall degradation performance is analyzed,and the relationship between the exponential model of wind turbine degradation and the actual degradation process is determined by analyzing the expectations of SDE.Secondly,SDE is used to simulate the actual state change of the wind turbine,and the exponential model is used to simulate the predicted state of the wind turbine.According to the martingale theory,the two state change trends are analyzed and it is determined that CBM is an optimization based on TBM.Finally,based on the update process theory,the influence of different maintenance time on the availability of wind turbines is analyzed,and then the CBM time is calculated with the availability constraints.(3)For multi-component systems,research the optimization method of condition-based opportunistic maintenance.Firstly,the CBM and degradation process of a single component are described by different SDE.By analyzing the expression of the optimal state of the component,the influence of various parameters on the state of the component is determined.Secondly,the method for obtaining the condition-based opportunistic interval of multiple components is analyzed,and finally the optimal condition-based opportunistic maintenance cost is calculated according to the optimization algorithm.(4)Take the fault data of a wind farm as an example to simulate and analyze the CBM optimization principle.The cost analysis of TBM and CBM verifies that CBM of wind turbines is an optimization of TBM,and reasonable CBM can improve state stability.Through the analysis of the condition-based opportunistic maintenance cost,it is verified that the condition-based opportunistic maintenance of multiple components further reduces the CBM cost. |
