Reliability Analysis And Maintenance Optimization For Wind Turbine Based On SCADA Data

In recent years,China's wind power industry has developed rapidly,and the total capacity of installed wind power has topped the world.However,due to the late start of wind power industry and the backward technology in the early stage,there are many defects in the design and manufacture of the early wind turbines.As the warranty of these early wind turbines terminates in succession,the problem of reliability is becoming more and more prominent,which seriously restricts the development of China's wind power industry.Based on the field operation data acquired from a wind farm in Jiangsu Province,the reliability characteristics of domestic wind turbines are analyzed,then the single part and the multi-component maintenance optimization problem are studied.The main works of this paper are as follows:(1)Taking a batch of domestic wind turbines as the object,data mining techniques are used to analyze the SCADA data of these wind turbines.The reliability of wind turbines is evaluated by calculating the failure rate,time between failures and availability.The results are compared with the research abroad.On that basis,time series method is used to study the relationship between the environment temperature and the failure rate of wind turbines.It shows that wind turbines have a higher failure rate in hot and cold weather,and their reliability has obvious seasonal feature.(2)The single part maintenance optimization of wind turbines is studied.The influence of minimum maintenance,imperfect maintenance and replacement on the effective age of the component are analyzed.The replacement cycle of the component is divided into several maintenance periods.Taking the number of maintenance periods and the length of each maintenance period as variables,and by considering the maintenance cost,outage losses,fixed cost and availability,a maintenance optimization model for single part system is proposed to minimize maintenance cost per unit time in a replacement period.The interior point method and enumeration method are used to solve the model.A case study is provided to prove the validity of the model.The result shows that compared with the traditional periodical maintenance and sequential maintenance,the model can make a more reasonable division for maintenance intervals.(3)The multi-component maintenance optimization of wind turbines is studied.The maintenance thresholds are set for the effective age of components.Taking the maintenance thresholds as the variables,considering the influence of the seasons on the failure rate of the components,a maintenance optimization model for multi-component system is proposed to minimize wind turbine's maintenance cost during the design life.Genetic algorithm is adopted to solve the model,and the validity is verified with a case study.The result shows that this model can provide the optimal maintenance plan according to the inherent reliability and maintainability characteristics of the components.In this thesis,the reliability of wind turbines is evaluated based on the SCADA data,and the seasonal feature of the reliability of wind turbines and their components are analyzed considering the influence of the environment temperature.Then,taking into account the availability and the growth of the effective age,the single part sequential maintenance optimization of wind turbines is studied.Finally,considering the influence of the seasons,the multi-component opportunity maintenance optimization of wind turbines is studied.The above researches focus on wind turbine performance evaluation and maintenance decision-making,which can provide a theoretical basis for wind farms to make scientific maintenance plan.