Algebraic estimation-based model-free control for variable-speed wind turbines

This research study validates the recently invented Model-free Control (MFC) approach via algebraic estimation to a variable-speed wind turbine. Instead of counting on adaption rules or neural networks, this novel technique employs an ultra-local model for rapid algebraic parameter estimation, requiring no statistical expertise about system models, disturbances, or interdependence between inputs and outputs. In low-speed winds, a model-free controller is designed based on a wind estimator to track the optimum value of power coefficients. In high-speed winds, MFC is extended to a multi-variable model-free controller (MV MFC) in pitch and torque to regulate the produced power around its nominal value. The approach’s key advantage lies in eliminating complex non-linear models, ensuring stability, and reducing non-linearities, uncertainties, and perturbations. The controllers have been tested with a wind turbine mathematical model and validated with FAST simulator in the presence of measurement noise and disturbances. The results show the proposed controller’s superiority over the existing techniques and improvement of power capture. • Validation of Model-Free Control (MFC) approach using algebraic estimation to variable speed wind turbines. • In low-speed region, A model-free controller based wind estimator is proposed to maximize energy capture. • In high-speed region, MFC is extended to a multi-variable model-free control (MV MFC) to maintain a satisfactory power quality. • Elimination of complex system modeling of the wind turbine, enabling real-time updates without concerns about linearity or specific parameter identification. • The proposed controllers lead to an increased performance level of the wind turbine operation compared to existing control strategies.