Posters

Abstract

Considering the atmospheric thermal stability conditions is a prerequisite to perform accurate WRA based on computational fluid dynamics (CFD). Wind resource assessment based on thermal stability consists first of determining the thermal stability conditions of the wind farm area, which can be stable, neutral or unstable with daily and seasonal variability. Then, depending on these site conditions, advanced CFD tools can adapt RANS equations and boundary conditions to perform thermal stability-dependent simulations. The application of a general method to determine thermal stability conditions is still a challenge for developers. The most direct method to determine the atmospheric stratification aims at estimating the Monin-Obukhov length or Richardson numbers using directly the heat flux or temperature profiles based on Monin-Obukhov similarity theory (MOST). However, for engineering purposes, using this approach can be difficult due to poor data availability or quality on site. In previous work, alternative methods have been developed for simple and complex terrain based on standard met mast measurements or mesoscale data. The aim of this work was to evaluate the precision of these determination methods for CFD simulations. The main challenge on this project was to investigate if negative wind shear or “S-shaped profiles”, non-visible using neutral simulations, could be simulated properly by the CFD with specific thermal stability conditions. S-shaped profiles can be considered as non-idealized wind profiles and can affect not only power production but also the loads on the structure. For this purpose, two sites have been investigated containing several masts with common measurement periods and heights. The relevancy of several determination methods used to deduce stability conditions will be discussed. We will show that an appropriate selection of the thermal stability conditions enables accurate simulation of S-shape profiles but also significant improvements in the wind speed cross-predictions.