Adaptive Observer Based Fault Diagnosis For Wind Turbines

With fossil fuel consumption increasing and environmental problems becoming serious,the world energy supply has relied on renewable energy.Among them,wind energy is wide application in various countries in recent years.However,wind farms are usually located in desolate and remote areas.Wind turbines operate in a harsh environment for a long time,leading to mechanical faults,electrical faults.It is necessary to adopt an effective wind turbine fault diagnosis method to timely detect hidden faults during operation time.As a result,the operating reliability and economic benefits of wind turbines can be improved.The wind turbine benchmark model is adopted in this research.This model divides the wind turbine into four components:blade&pitch system,drive train,generator&converter,and controller.The sensor faults,actuator faults,and system faults are considered.To determine whether a fault has occurred and where the fault is located,an adaptive observer based methods is adopted.This method constructs a system model based on the initial state of the system and control signals.The residuals are obtained by comparing the predicted value with the actual output value,thus fault can be detected by analyzing the residuals.This method has mature technology and strong feasibility thus received widespread attention.As a model-based fault diagnosis method,the adaptive observer scheme can obtain the estimated value of the system state and output.It can also estimate the fault value,which is convenient for the subsequent design of the fault-tolerant controller.An adaptive observer-based fault diagnosis method for the wind turbine is proposed in this research.Firstly,the subsystems of the wind turbine benchmark model are reconstructed.Based on the faulty models,the corresponding adaptive observers are designed to estimate the system states and faults.The residual is calculated by subtracting the actual output value of the benchmark model from the estimated value obtained by the observer.To improve the effectiveness of the observer,we adopt the Fast Adaptive Fault Estimation(FAFE)algorithm in the design to estimate the fault.Compared with the conventional one,the fault estimation performance of this algorithm is faster and more accurate.According to the interference and noise in the wind speed,the state-space model of the drive train system is expressed as nonlinear.Therefore,the order of the drive train state-space model is reduced to eliminate the influence of disturbance.Besides,an adaptive observer based fault-tolerant control scheme is proposed for the blade&pitch subsystem and the generator&converter subsystem,thus reducing the impact of faults on the system and maintain stable performance for the wind turbine under fault conditions.Based on the estimated information obtained by the adaptive observer,a state feedback fault-tolerant controller is designed.This research conducts simulation experiments in the MATLAB/Simulink environment,selects data within 4400s.Sensor faults,actuator faults,and system faults are considered.The results show that the faults can be located by different components’ residual.The scheme proposed in this article can timely detect and isolate the faults.Besides,the fault-tolerant controller designed for the blade&pitch subsystem and the generator&converter subsystem has been experimentally proved to effectively stable the system during the faulty period,and the performance is the same as the normal state.Therefore,the proposed scheme can tolerate the considered faults on these two systems.Furthermore,the proposed method has better performance both in fault estimation and fault tolerance compared with the existing scheme.