ECCOMAS 2024

Homogenized Lattice Boltzmann Methods for City-Scale Flow Simulations in Digital Twins

  • Teutscher, Dennis (Karlsruhe Institute of Technology)
  • Bukreev, Fedor (Karlsruhe Institute of Technology)
  • Simonis, Stephan (Karlsruhe Institute of Technology)
  • Krause, Mathias (Karlsruhe Institute of Technology)

Please login to view abstract download link

We present a coherent methodology for city-scale flow simulations using homogenized lattice Boltzmann methods (HLBMs) implemented in the open-source software library OpenLB [2]. Based on approximating the governing partial differential equation system (including homogenized Navier–Stokes equations and advection–diffusion equations ), we enable a comprehensive modeling of flow dynamics, particle distributions, and special objects of the urban environment considered as porous media. As a pivotal component of our methodology, we incorporate publicly available meteorological data sourced from monitoring stations. This integration not only enhances the accuracy and authenticity of our simulations but also facilitates the development of a sophisticated digital twin for the city under consideration. The digital twin serves as a virtual replica, providing an invaluable tool for in-depth analysis and exploration of the urban wind flow dynamics. Combining mathematical modeling, consistent numerical methods and efficient and scalable implementations, our novel approach allows for computational analysis of the influence of small objects (e.g. trees) on city-scale wind flow, and paves the way to data-based urban design strategies.