INTRODUCTION
The resurgence of wind as a viable energy resource has resulted in a large volume of investment in development of wind turbines. To support the turbine design, various computational models have been developed to determine wind flow patterns over terrain. The goal of such models is to maximise turbine efficiency by finding sites with maximum air flow. One of the more sophisticated methods has been developed by the School of Engineering of Edinburgh University, which is based on the Weather Research and Forecasting (WRF) system. This model does an excellent job in representing hourly time series at a 3km resolution of UK and Irish wind resources; however its performance in mountainous and coastal areas is less clear, due to a more complex behaviour of the atmospheric boundary layer. This difficulty arises due to the strain associated with computing fluid flow on a large scale; leading to low resolution results which cannot account for steep changes in gradient of the terrain and temperature as well as sudden changes of surface type. Additionally the WRF uses a linear system of equations and cannot account for the more complex turbulent flow associated with such features.
It is therefore the aim of this project to add to the sophistication of the current model by coupling a finer microscale analysis, using a non-linear CFD model, to the mesoscale predictions. Such a model will allow for a more precise positioning of wind turbines within the selected development site.
It is therefore the aim of this project to add to the sophistication of the current model by coupling a finer microscale analysis, using a non-linear CFD model, to the mesoscale predictions. Such a model will allow for a more precise positioning of wind turbines within the selected development site.