A CFD-GENERATED WIND FIELD LIBRARY FEASIBILITY STUDY: MAXIMUM WIND DIRECTION INTERVAL

摘要

Computational fluid dynamics (CFD) models are capable of computing 3D flow patterns around arbitrary building arrangements and have shown promise for building-to-urban scale transport and dispersion problems (e.g., Murakami, 1997; Brown et al., 2001).However, their utility for planning and assessment applications has been limited due to computational requirements, expertise requisites, and relatively long turn-around time. The possibility of using a library of pre-computed mean wind and turbulence fields for a prescribed urban site has been suggested as a way of circumventing the usual complaints, while simultaneously providing more detailed and accurate information to planners than simpler urban dispersion models. In this paper, we will present afirst step in evaluating whether or not this approach is feasible, i.e., examine the computing and storage requirements to build a wind library. To populate such a database, one needs to specify the prevailing wind direction interval. Using Lower Manhattan as a test case, we will try to determine the wind direction interval required to adequately describe flow patterns in a downtown area for transport and dispersion problems.