New perspectives on maximum wind energy extraction of variable-speed wind turbines using previewed wind speeds

Abstract

Maximum wind energy extraction remains a challenge for the large-scale wind turbines due to their increasing moment of inertia. In this study, a novel nonlinear predictive control method using previewed wind speeds is proposed to address the issue of maximum wind energy extraction for variable-speed wind turbines. Different from the conventional structure, the proposed nonlinear predictive control is based on the discrete prediction model established by a large time step. By doing so, the number of time discrete points is significantly reduced and the quantity of the solution set becomes numerable. As a result, the implementation complexity and computational burden are able to be relieved, and the real-time implementation becomes practical and easy to carry out. The proposed method is demonstrated on two models of wind turbines with different scales, and its performance is compared with the industrial controller. Comparison results show that the increased energy extraction efficiency gained by the proposed method is more than 1% on the two turbine models under different wind conditions, and thus the superior capability in extracting wind energy promises potential industrial applications. Meanwhile, the expense is the considerable increment of torque variation, which can be relieved by revising the control parameters. The presented method provides new insights into the design of the control methods for wind turbines equipped with advanced measurement devices.