Development of designer-relevant Lattice-Boltzmann Wind Field Model for urban canyons and their neighborhoods

Abstract

Wind field analysis is one of the most important components for designers to achieve a thermally-comfortable and energy-efficient building design. Designers need a fast and relatively accurate wind field model to get integrated into the design workflow, but current platforms to work on are either costly and time-consuming conventional Computational Fluid Dynamics (CFD) tools or over-simplified data correlation factors, which makes the workflow undesirable for designers' use. In this thesis, a novel Lattice-Boltzmann Wind Field Model (LBWFM) is developed and integrated in a designer-relevant Rhino-based environment. Lattice-Boltzmann Method (LBM) is introduced as the solver due to its open-source and parallelism natures, and coded in C# language for three-dimensional urban airflows. Results of the model are validated with experimental measurements as well as conventional CFD tools for both wind velocity and pressure fields. To further enhance the computational efficiency, proper settings of inlet wind profile and optimal modeling domain size are investigated for the LBWFM. And the relative wind pressure coefficient calculated out of the model is then applied in the analysis of wind-driven natural ventilation potential with the indicator of air exchange flow rate. Finally the limitation of the model is stated and future work is discussed on the modifications of buoyancy effect and potential extension is addressed in the application of LBWFM.