| Modern wind turbines are developing in the direction of large-scale and complex.Due to the complex and harsh environment with the complex and changeable wind loads,superposed shocks of ocean waves,high temperature,high humidity,aggressive salt corrosion,large offshore wind turbines are confronting daunting challenges of the service reliability issue.The investment risk of offshore wind power is high,and the cost of operation and maintenance is much higher than that of onshore wind power.Hence,the higher requirements are placed on the reliability of offshore wind turbines.Improving reliability becomes the cornerstone of the development of offshore wind power.Most of the existing reliability research on wind turbines focuses on components and assemblies,and the reliability status of components and assemblies can only be qualitatively analyzed,such as finite element analysis,dynamic characteristic analysis,etc.There is a lack of quantitative assessment of the time-varying reliability of the system throughout their life cycle.The failure modes and failure mechanisms of various subsystems of large wind turbines are different,such as the hydraulic system,the drivetrain system,and the electrical system.It is necessary to study the structure and failure mechanism of the system and fully consider the structural characteristics and failure modes of the system.The multidisciplinary approach is adopted to establish the time-varying reliability model of each subsystem.The large wind turbine is a complex electromechanical-hydraulic integrated equipment with multi systems.The complexity,technical level,importance,and working environment of each component are quite different.To perform the reliability allocation of critical components,it needs to fully consider the influential factors and service time of components and distribute the reliability index of the whole wind turbine proposed by customers to each component and subsystem.This thesis takes a large wind turbine as the research object,statistically analyzes the field reliability data,determines the critical subsystem of this type of wind turbine,and analyzes the failure mechanism of critical subsystems.For the hydraulic system,drivetrain system and redundant system of the wind turbine,based on the structural characteristics and failure mechanism of the system,considering the load-sharing characteristics of components,fatigue damage,strength degradation,component swapping,and imprecise parameters,time-varying reliability modeling and evaluation methods for critical subsystems are explored.The influences of relevant factors on the subsystem’s time-varying reliability are revealed.Besides,the reliability importance and reliability sensitivity of critical components are studied.According to the field reliability data of wind turbines and the time-varying reliability evaluation of critical subsystems,comprehensively considering the influential factors of critical components and the service time,the improved reliability allocation method is adopted to perform the reliability allocation of critical components.The evaluation of the allocation results is conducted,which can provide theoretical support for improving the reliability and availability of large wind turbines.The main research contents of this thesis are classified as follows:(1)After discussing the problems existing in the management of field reliability data of wind turbines in China,some opinions and suggestions are given.The critical subsystems of this type of wind turbines are determined through statistical analysis of the fault data of wind farms.Following this,the failure mechanisms of critical subsystems are analyzed,which can provide conditions for system time-varying reliability modeling.(2)According to the structural characteristics and failure modes of the wind turbine hydraulic system,considering the load-sharing characteristics of redundant components and the imprecise failure information,the time-varying reliability modeling method of the wind turbine hydraulic system is studied.The dynamic fault tree-based time-varying reliability model and the survival signature-based time-varying reliability model of the hydraulic system are established.Based on the system’s reliability models,the fuzzy failure probability of different states of the hydraulic system at a given time,and the fuzzy reliability at a given membership degree are analyzed.Following this,the effects of different modeling methods of load sharing on the system reliability are explored.(3)The time-varying reliability modeling and evaluation method of the wind turbine drivetrain system based on fatigue damage and strength degradation is studied.An aero-servo-elastic coupling dynamic model of the large wind turbine is established.Based on this,the load-time history of critical transmission components can be obtained.Following this,the calculation formula of the contact stress of gears is derived.The average-amplitude matrixes of the bending stress and contact stress of gears are obtained using the rain flow counting method,whose probability density distribution functions are tested and confirmed.The trapezoidal fuzzy number is used to represent the strength degradation process of components,and the time-varying reliability functions of gears and bearings are derived.The load-sharing characteristics of planetary gears and downwind dual bearings are considered,and the time-varying reliability model of the wind turbine drivetrain system is established based on the fault tree of the drivetrain system.Besides,effects of different wind conditions on the time-varying fatigue reliability of the bearing,gear,and drivetrain system are explored,and influence rules of the strength degradation on the reliability and failure rate of the drivetrain system are revealed.(4)Based on a comprehensive and deep analysis of the multiple failure mechanisms of the redundant system of large wind turbines,the reliability function of the redundant system is derived.The survival signature is used to represent components’ swapping,and the time-varying reliability model of the redundant system of wind turbines based on component swapping is established.Following this,the effects of different swapping strategies on the time-varying reliability of redundant systems are explored,and change laws of the time-varying reliability of the wind turbine redundant system are revealed.Besides,P-Box is used to describe the imprecise failure rate of components.The marginal reliability importance and the joint reliability importance of redundant components are studied.This research can provide new technical means for the component swapping-based reliability improvement of large wind turbines.(5)The influencing factors and the service time of critical components are fully considered,and an improved reliability allocation method is proposed,which can allocate the system reliability index proposed by customers to each component.Moreover,the allocation results obtained by the improved method are compared with that of traditional allocation methods.Combined with maintenance allocation and prediction of components,the reliability evaluation method based on the compound weight and fuzzy numbers is used to evaluate the reliability allocation results.The reliability of weaknesses is improved through redundant design,and the system’s time-varying reliability evaluation model is established.Finally,the closed-loop reliability design of large wind turbines is realized. |