A Robust Adaptive Extremum-Seeking-Based Optimal Torque Curve Tracking for Wind Turbine Generators

Optimal torque curve control is a common technique used to track the maximum power point of wind energy systems without direct wind measurements. However, it relies on precise knowledge of the turbine's aerodynamic characteristics and air density. Since these parameters can differ significantly from their nominal value due to variable ambient conditions and aging of the turbine, suboptimal operation of the wind generator can occur. In this paper, a robust and adaptive Extremum Seeking optimization to track the optimal torque trajectory and achieve maximum wind energy harvesting is proposed and implemented. Unlike other approaches found in the literature, adaptive Extremum Seeking is leveraged here to drive the generator torque toward its optimal trajectory rather than to define a variable speed set-point for the turbine. By doing so, maximum-power-point operation can be achieved with reduced oscillations in torque and electrical power. Furthermore, the detection of wrong derivative estimates is integrated into the proposed algorithm to acquire robustness against sudden wind changes, which may otherwise compromise tracking stability and performance. The results of simulations on a 1.5 MW wind energy system and extensive experimentation on a small-scale test bench are presented to demonstrate the efficacy of the proposed technique.