An Urban-Conscious Rapid Wind Downscaling Model for Early Design Stages

Assessments of urban contexts using existing microclimate models mostly fall short, when considering topographies along with complex layouts of buildings and streets, regardless of their significant influences on building performances and outdoor environments. The challenge exists mainly due to modelâ??s inherent complexities and the associated high computational costs. This becomes especially challenging at early design stages when time, expertise, and computational resources are limited, even though the opportunities for performance enhancement are greater than at later stages. This dissertation develops a wind downscaling model that can rapidly assess urban contexts to relate climate data in a large spatial resolution for a smaller-scale site. Surrounding slopes and terrains, up to a few kilometers in diameter, are considered to predict wind pressure on the volumetric boundary of a neighborhood and local wind speed. The new model strives for prediction accuracy and computational efficiency by employing the capacities of a computational fluid dynamics (CFD) simulation and of an existing mathematical method. The proposed model is composed of three parts: pressure database, speed database, and interpolation. The databases store wind data for existing urban contexts that are generated with CFD simulations. Using the databases, the interpolation approximates the pressure outcomes for a new urban context; thus, real-time CFD runs can be avoided for the model users. Independent development of data for pressure and speed facilitates the flexibility and expandability of the model. The proposed model showed an acceptable prediction accuracy, with average errors of less than 10%, compared to the full-scale CFD simulation for the same territorial scope. An exceptional computational efficiency is also shown, with a runtime in 0.308 seconds, which is 16568 times faster than the CFD simulation. This rate allows creation of a yearlong prediction in a few tens of minutes with a personal desktop computer. For non-experts, the pertinence of the model is enhanced with a limited number of parameters, making it easily adaptable during early design stages of buildings and urban design scales. Geometric sensitivities are embedded for incremental study, which is crucial to finding optimal solutions, toward more efficient, yet healthier, urban environments. Degree Type Dissertation Degree Name Doctor of Philosophy (PhD) Graduate Group Architecture First Advisor Ali M. Malkawi This dissertation is available at ScholarlyCommons: http://repository.upenn.edu/edissertations/1076 Second Advisor Yun K. Yi