A paper published today by researchers at National Renewable Energy Laboratory in Colorado outlines a novel approach to optimising wind power plant layouts using artificial intelligence (AI) techniques. It emphasises the importance of accommodating geodiversity in wind power development to support its expansion across diverse geographic locations. The proposed method focuses on implementing wake steering, a control strategy that directs turbine wakes to optimise energy production at wind power plants.
The study evaluates the potential benefits of wake steering at over 6,800 plausible onshore wind locations in the USA. By employing AI models trained on engineering wind flow simulations, the researchers estimate the impact of co-optimising plant layouts with wake steering on land requirements and economic benefits. On average, the implementation of wake steering could reduce land requirements by 18% per plant, with site-specific benefits ranging from 2% to 34%. However, these estimates are subject to uncertainties stemming from factors such as the wind flow model, wind resource estimates, buildout scenarios, and geographic variations.
Furthermore, the study predicts that wake steering could enhance power production during high-value periods, leading to potential revenue increases of up to US$3.7 million per plant annually, equivalent to US$13,000 per megawatt per year. Nevertheless, the economic gains may vary significantly depending on specific settings.
Overall, the consideration of wake steering’s geographic potential highlights varying nationwide prospects for improved economics and siting flexibility in wind power development. This research underscores the importance of innovative approaches and advanced technologies, such as AI-driven optimisation, in supporting the growth of clean energy infrastructure.
Source
Artificial intelligence-aided wind plant optimization for nationwide evaluation of land use and economic benefits of wake steering, Nature Energy, 2024-04-22
Boˆreal 