Observer-based Fault-tolerant Control Of DFIG-based Wind Energy Conversion Systems

This thesis proposes four fault-tolerant control(FTC)strategies based on observers for a doubly-fed induction generator(DFIG)-based wind energy conversion system(WECS)with sensor faults.The contributions of this thesis are summarized as follows.This thesis presents a Kalman filter-based fault-tolerant control(KFFTC)strategy for a DFIG-based WECS under voltage and current sensor faults.Based on the independent timevarying models of stator voltages,stator currents,and rotor currents,six Kalman filters(KFs)are designed in parallel to estimate voltage and current components in the presence of measurement noise.The sensor faults are detected and isolated based on the residuals calculated from observations obtained by sensors and estimations provided by Kalman filters.The faulty measured signals are then replaced by the estimated signals derived from corresponding Kalman filter to reconfigure the control system of DFIG during the sensor faults.Simulation studies undertaken on a grid-connected DFIG system reveal that the KFFTC strategy is able to correctly detect the sensor faults and isolate the faulty sensor,and it ensures the fault-tolerant operation of DFIG under the conditions of stator-voltage,stator-current,and rotor-current sensor faults.This thesis presents an improved fault-tolerant control(IFTC)strategy for a DFIG-based WECS subject to rotor and stator current sensor faults.A novel stator-current-loop vector control(SVC)scheme is proposed to realize the decoupled control of active and reactive powers of DFIG without involving rotor currents,and it is compared with the conventional rotor-currentloop vector control(RVC)scheme on tracking performance,closed-loop stability,and robustness against model uncertainties and external disturbances through theoretical and simulation analysis.The IFTC strategy is obtained through replacing the RVC of KFFTC strategy with novel SVC,thus improving the sensor fault tolerance of DFIG-based WECS against current sensor faults.This thesis presents a current sensor fault-tolerant control(CSFTC)strategy,combining perturbation observer-based direct power control(PODPC)and two-stage Kalman filter(TSKF),to enhance the fault tolerance of a DFIG-based WECS with rotor and stator current sensor faults.In the PODPC scheme,the interactions between active and reactive power control loops are represented by newly introduced perturbation states,and the feedback linearization control is realized with the state estimations derived from perturbation observers to achieve the decoupled power control.Neither rotor current sensors nor parameters of DFIG are required in the implementation of PODPC.Stator current TSKFs are designed to generate residuals for fault detection and isolation,and provide current estimations to replace the faulty current measurements for system reconfiguration under stator current sensor faults.Simulation studies undertaken on a grid-connected DFIG system reveal that the proposed CSFTC strategy is immune to rotor current sensor faults,and it provides strong fault tolerance to stator current sensor faults.This thesis proposes a fault observer-based fault-tolerant control(FOFTC)strategy for a DFIG-based WECS with sensor faults.The DFIG system is decomposed into stator,rotor,grid current models,and the sensor fault observer(SFO)is presented to provide accurate estimations of sensor fault components in the presence of model uncertainties.Based on the independent current dynamics of DFIG system,three SFOs are designed in parallel to respectively estimate the stator,rotor,and grid current sensor fault signals.During the occurrence of faults,the fault estimations derived from current observers are adopted to compensate the faulty signals,thereby achieving the system reconfiguration of DFIG without requiring any fault diagnosis schemes.Simulation studies are undertaken on a grid-connected WECS under the conditions of stepped and random wind speeds,and single and multiple current sensor faults.The simulation results reveal that the FOFTC controlled DFIG system is resilient from the sensor faults and maintains continuous stable operation.