Research On Fault Diagnosis And Fault Tolerant Control Of Wind Turbine System

Due to the prominent issue of energy shortage and environmental constraints all over the world, the demands of renewable energy is increased, and wind turbine technology has been rapid development accordingly. Meanwhile, fault diagnosis and fault-tolerant control (FTC) research of wind turbine is gaining more and more attention, with the requirement of the reliability of power generationWind power generation process is a complex process. This paper summarizes the previous work, the models of key components of wind turbine system which contains aerodynamics system, pitch system, drive train system and generator system are established respectively. Then the dynamic model of the whole system can be obtained by integrating the subsystems which provide the basis for next studies.This paper presents a fault diagnosis method based on parameter identification for wind turbine system. Recursive least squares method is adopted to estimate the model parameters of wind turbine system. Bayesian decision theory is used to analysis the parameters changes, then the fault can be detected. The type of fault is determined using neural network through process the failure information. Simulation results show the method can diagnosis system failure rapidly and accurately.A multiple model predictive control based FTC strategy for wind turbine is designed in this paper. The multiple models of wind turbine system in different condition are presented and the predictive controllers are designed according to each system model. When fault occurred, the corresponding model is switched to achieve the purpose of fault tolerance. The test is simulated in the environment of MATLAB, and the results validate the feasibility and effectiveness of the proposed FTC strategy.