Multi-state Reliability Assessment And Maintenance Policy Optimization Of Wind Turbines

With the rapid growth of energy demand and the deteriorarion of the environmental crisis,the energy consumption structure of China which concentrates on coal and oil seriously is not suitable for the sustainable development.As a clean and environment friendly renewable energy,wind energy has attracted widespread attention.Althougth,wind power generation technology is mature and low-cost,but there are still some problems such as high failure rate,not meeting design life requirements,and the high cost of wind farm operation and maintenance,which restrict the development of wind power generation.Therefore,based on the small sample data of wind turbine,this paper carries out the work of the following two aspects,reliability analysis and maintenance strategy optimization of mechanical components.In the first part,aiming at the problem that large wind turbine faces the dilemma of insufficient statistical data,in this work,an improved Hidden Markov Model reliability evaluation method considering performance degradation is proposed,called degradationhidden-markov model.The method derives the component performance degradation rule based on monitoring data of the operating parameters,and hidden markov model is improved by performance degradation rule to obtain a new time-correlated state transition probability matrix with degradation characteristics.The state probability and reliability curve are obtained by degradation-hidden-markov model,thus the reliability analysis of wind turbine components based on small sample data is realized.Based on the performance data of wind turbine generator,the state probability and reliability evaluation results of the Hidden Markov Model and the degradation-hidden-markov are compared.The results show that the reliability curve calculated by the degradation-hidden-markov model is more consistent with the degradation trend.In the second part,due to the high failure rate of the wind turbine,the wind turbine operation and maintenance cost is high.This paper establishes a opportunity maintenance strategy optimization model based on multi-state reliability.According to the improved hidden markov model,the time-state probability-reliability graph of the wind turbine component is obtained,and then the performance data is used to analyze the operating state of the component and predict the reliability data.According to the reliability curve,the impact of the failure risk on the maintenance cost is quantified.Thus the pre-maintenance total cost model considering the risk cost is established.The condition-based maintenance is guided according to the state probability and reliability of the component in performance degradation period,the preventive maintenance period and the opportunity maintenance period are defined as well.Therefore,the preventive maintenance cost and opportunity maintenance cost over the life cycle are calculated.With the delay of the maintenance time,the impact on the maintenance cost caused by the single maintenance cost and maintenance cycle is analyzed.Aim at the lowest maintenance cost,the optimal maintenance strategy can be determined.