HIDALGO COE and ENCCS Webinar: Use case with SLB Analys

The HiDALGO2 Centre of Excellence (CoE) and EuroCC National Competence Center Sweden (ENCCS) in cooperation with CASTIEL 2 are pleased to offer a webinar on May 26 at 13.00 CET with the title: HiDALGO2 COE and ENCCS: Use case with SLB Analys

The webinar will be held over Zoom. To register, click here.

Abstract: High-resolution urban wind comfort computation for the entire city of Stockholm

Authors: Zoltán Horváth, Mátyás Constans, Széchenyi István University (SZE)

*Figure 1: Original building data provided by City of Stockholm, with OpenStreet Map’s dataset rendered below. Copyrights: OpenStreet MAP for the data rendered above, SZE for the overall image.*

The challenge

The scientific challenge of the Castiel2-webinar is to provide wind-comfort information by computational simulations for the entire city of Stockholm, i.e. to compute wind speed at each point of the city under various weather conditions.

The point is that wind speed near tall buildings might be much larger than the average speed, putting pedestrians at risk. The wind-comfort information, in particular when applied to several planned building scenarios, can be used effectively by the city planners to optimize next constructions.

The requested resolution of the full computational domain of 100 km2 in space is 1-2 metres at ground level resulting in the computation of tens of millions of variables for each scenario, which requires HPC.

The diverse building and terrain geometry of the city have to be treated properly, resulting in a very complex mesh generation process. A major challenge is visualizing the resulting data, which can be several terabytes in size, via a web browser while maintaining a smooth user experience.

The solution

To address the scientific challenges, we apply the HiDALGO2 Urban Air Project (UAP) solution, developed by Széchenyi István University (SZE), for Stockholm.

The external stakeholder SLB-analys, a unit of the Environment and Health Administration of the City of Stockholm provided data for the simulations through a collaboration with ENCCS, the EuroCC National Competence Center Sweden.

In the webinar the UAP solution will be presented.

The pillars of the solutions are as follows:

Mesh generation from the shape files,
CFD solution on EuroHPC and local machines by the native multi-GPU code RedSim,
CFD-results on HPC are visualized by CFDR, which can render CFD-related 3D graphics, in real time, on a web browser (even on low-end laptops).