Comparing WAsP and MeteodynWT estimates for the 'regular' user

摘要

The application of Computational Fluid Dynamics (CFD) for wind resource assessment has become a real alternative to common linear model approaches, like WAsP. However, the increase in complexity and user expertise in CFD calculations leads to greater difficulty when comparing performance between modeling approaches. Trying to contribute to some enlightenment on these issues, namely in what regards relative performance between WAsP and Meteodyn CFD code estimates, the authors have gathered a set of 10 test cases in which model performance was compared. Cases are diverse in terms of location, wind climate, land cover, although all of them are in complex to extremely complex sites. To set aside dependence on user expertise, the authors have outlined a standard and uniform approach to CFD model setup, regardless of the case at hand. Model performance was evaluated by means of estimated average wind speed for sites where real observations were available. Errors from CFD estimates were significantly lower than those from WAsP in 9 out of 10 cases. Excluding the worst test case, average RMS from CFD is lower than half the RMS from WAsP. Scatter around RMS is still high, making a solid conclusion still risky. The authors would like to state that, regardless of the good CFD performance shown by this results, it is still not safe to establish that this CFD provides a better solution to wind resource assessment when compared to WAsP. The total number of cases is still small and the diversity of site conditions is far from appropriate for a more general conclusion. Moreover, cross-prediction of average wind speed for masts sites and measurement heights may not provide the best way to fully assess performance of a resource assessment and micrositing tool.