An innovative hybrid grid generation for urban ventilation simulation

Urban ventilation study using the computation fluid dynamics (CFD) approach plays an important role in alleviating the urbanization challenges of high energy consumption, human comfort, and public health in an urban built environment. Grid arrangement is a critical step for such CFD applications. The grid amount and quality can significantly influence the accuracy, convergence, and computational efficiency. In this study, an innovative hybrid grid generation strategy is proposed for urban ventilation simulation. This strategy aims to strike a balance between capturing the intricate details of the flow and minimizing the computational burden associated with a large number of grids. The proposed hybrid grid algorithm characterizes an octree-based Cartesian grid dominating the computational domain, mixing with a tetrahedral grid close to the building surfaces. The performance of this hybrid grid scheme is tested with an example of a single high-rise residential building, and a significant reduction of up to 98% in the number of grids is achieved for the hybrid grid scheme compared to the all-tetrahedral grid scheme. The multiple buildings example shows that this proposed grid generation can expand to general urban-scale built environment applications. Finally, a case study with the hybrid grid arrangement is carried out to validate the simulation results with experimental measurement data. The proposed hybrid grid generation in this study offers an efficient and reliable approach to simulate urban ventilation by an automatic grid generation process. This highly efficient grid arrangement can shorten the design turnaround and contribute to integration into big data for urban built environments with efficient data storage and data processing.