A Computational Fluid Dynamics based framework to assess the wind energy potential of an urban landscape: A case study in Brussels

Raghunathan Srikumar Sampath Kumar Gabriele Mosca Ioannis Tsionas Llaguno-Munitxa, Maider [UCL] Stephan, André [UCL] Alessandro Gambale

Wind is a clean and renewable energy source that has the potential to significantly contribute to the electricity supply in urban areas. Electricity generation through Micro Wind Turbines (MWTs) in an urban setting is not often implemented given their expected low performance due to low wind speed. Wind speed is indeed generally lower in urban areas than in open, rural, and coastal areas: buildings and structures represent obstacles to the wind flow and reduce its velocity. But those same obstacles also locally accelerate the flow at some locations, so that accurate positioning of wind turbines can often result in a satisfactory performance. In the present work, a framework is detailed to assess the wind energy potential of an urban neighborhood using Computational Fluid Dynamics (CFD) and applied to the Northern District of Brussels, Belgium, a neighborhood that has the ambition to become a Positive Energy District. Assessing the wind energy potential of an urban area requires knowledge of local wind properties (speed, direction, turbulence) to a high spatial resolution, as conditions even on a single roof are not uniform. CFD is a powerful tool that can be used to discern wind patterns and aid in an accurate assessment of the wind energy potential. By using CFD, it is possible to accurately predict the wind speed, direction and turbulence within an urban landscape, taking into account the effects of buildings, terrain and other structures. Statistical wind data from the last 30 years collected by the nearest meteorological station is used to define the conditions for a large enough number (typically 5-10) of CFD simulations for each wind direction. Based on the obtained results, the potential energy output of a specific MWT can be predicted and sites with suitable conditions can be identified. Simulations are performed using the opensource finite-volume solver OpenFOAM v7. A modified RANS turbulence model (k-ω SST) with the improved Atmospheric Boundary Layer (ABL) approach from Bellegoni et al is used to solve the flow equations in order to improve the accuracy of results. The numerical analysis allowed to identify the most suitable locations for MWTs in Brussels Northern District, demonstrating how the described approach can be effectively used in assessing the wind energy potential in an urban environment.