1220 HPC application on turbulent wind over urban model represented by individual shape of buildings

  • Tetsuro Tamura, Tokyo Institute of Technology
  • Yasuaki Ito, Shimizu Corporation
  • Hidenori Kawai, Ochanomizu University

Recent development of High Performance Computation (HPC) has made it possible to carry out large-scaled simulation of turbulence in city and to use the sophisticated CFD techniques for actual building shape models with high resolution mesh system. Currently, the level of HPC ranges in feasible subject or deep understanding in wind-related issues. As an objective of maintaining structural safety of buildings, HPC can accurately evaluate the peak as well as the mean wind pressure distributions on the surface of a very complex building using Large Eddy Simulation (LES). LES can be used for the actual wind resistant design of various kinds of buildings and structures in cities. Also, LES can deal with the turbulent wind in urban area where atmospheric diffusion or heat island phenomena cause some environmental problems of human life. For the scientific aspect, some new statistical evaluation techniques are recently being developed in various fields. Here the data science technique in the big data system has been appropriately employed as a new estimation method of the predictive value and its uncertainty for wind-related issues. The combination of the engineering LES with the meteorological model is also getting attention in set up of the city impact by natural phenomena. Accordingly, based on the feasible stage and deeply understanding stage of CFD, this MS focuses on the following issues:
1. Extreme disaster estimation based on hybrid method of meteorological model and engineering LES
2. Fact clarification and proposal of countermeasures in structural and environmental problems
3. HPC result and its development to engineering design stage
4. Understanding essential features of urban turbulence using data science technique

© WCCM-APCOM 2022. All Rights Reserved.