Researchers from Politecnico di Milano, including Sebastiano Randino, Lorenzo Schena, Nicolas Coudou, Emanuele Garone, and Miguel Alfonso Mendez, have developed a new approach to model and control wind turbines, particularly those affected by wakes from other turbines. Their work, published in the journal Renewable Energy, focuses on improving the performance and stability of wind turbines in various operating conditions.
The team presents a nonlinear system identification framework that models the power extraction dynamics of wind turbines, both in freestream conditions and when they are affected by wakes from other turbines. The approach uses data-driven power coefficient maps, expressed as combinations of compact radial basis functions and polynomial bases, parameterized in terms of tip-speed ratio and upstream conditions. These surrogate models are then embedded in a first-order dynamic system suitable for model-based control.
The researchers validated their method through experiments in two wind tunnel configurations: a low-turbulence tandem setup and a high-turbulence wind farm scenario. In the tandem case, the identified model was integrated into an adapted Kω^2 controller, resulting in improved tip-speed ratio tracking and power stability compared to traditional blade element momentum (BEM)-based and steady-state models. In the wind farm scenario, the model effectively captured the statistical behavior of the turbines despite the unresolved turbulence.
The practical applications of this research for the energy sector are significant. By enabling interpretable, adaptive control across a range of operating conditions, the proposed method can enhance the performance and reliability of wind turbines, particularly in wind farms where turbines often operate in the wakes of others. This can lead to increased energy production and reduced wear and tear on the turbines, ultimately lowering the cost of wind energy. The method’s ability to adapt to different conditions without relying on black-box learning strategies makes it a promising tool for the wind energy industry.
This article is based on research available at arXiv.