Active Fault Tolerant Control of a Wind Turbine via Fuzzy MPC and Moving Horizon Estimation

Abstract This paper proposes an approach to active fault tolerant control (AFTC) of a variable-speed wind turbine subject to actuator faults. The actuator faults in the turbine pitch system and generator system are considered in this work. Takagi-Sugeno (T-S) fuzzy modeling of the combined driver train, pitch and generator systems is proposed to account for nonlinearities in the dynamics. Moving horizon estimation (MHE) based on T-S fuzzy modeling is proposed as the actuator fault estimation unit. Model predictive control (MPC) based on T-S fuzzy modeling is used in the design of the AFTC unit in which the predictive controller compensates the actuator faults and takes into account the model nonlinearity and turbine system constraints. A T-S fuzzy observer is used to estimate system states and thus an output feedback strategy of an observer-based predictive controller is formed.