Fault Tolerant Control of Wind Turbine Using Robust Model Predictive Min-Max approach
Résumé
Control and fault compensation in wind turbines represents a challenge due to the stochastic nature of the driving force of power generation in these systems, their inherent nonlinearity and high coupling between the control inputs. In this paper, robust nonlinear predictive control is considered in order to satisfy the control objectives in the full load regime. Model predictive control allows taking into account constraints on the inputs and outputs as well as model nonlinearity. Robust MPC allows taking into account parameter uncertainties, that are supposed in this work to be due to pitch angle actuator fault. A multi input multi output controller is developed to maintain the generated power at its nominal value and reduce the loads on the torque system. The controller is found to react actively to the fault and maintain the desired performance of the system by an increased solicitation of the non-faulty actuators.