Reduced-Order Modelling for Gas Dispersion Across Urban Areas in Digital Twins

Kurzfassung

In today’s modern societies, critical infrastructure is crucial, and ensuring public safety during potential disruptions is a top priority. Recent advancements in the field of digital twins (DTs) have transformed DTs into reliable tools in analysis and forecasting. However, creating an accurate virtual representation of infrastructure requires a deep understanding of the physical system and its dynamic behavior. With increasing system complexity, it becomes more and more difficult to obtain results of high-fidelity physics-based models within an acceptable response time, ideally in real-time. Fast-to-evaluate models are especially important in crisis situations or to explore multiple what-if scenarios in parallel. To this end, model order reduction methods can offer a promising solution. A workflow towards a DT for real-time prediction of airborne contaminant transport in urban environments - featuring a reduced-order model for moderate Reynolds numbers - was presented in [1]. This contribution further extends the reduced-order model and provides a thorough comparison of POD-Galerkin [2,3] and PODI [4]. Moreover, a PODI-based approach is presented that reduces the combined problem consisting of the incompressible Navier-Stokes equations and the advection-diffusion problem simultaneously. To facilitate scenario analysis, an interactive dashboard has been implemented to visualize contaminated areas within an urban domain based on user-defined input parameters. REFERENCES [1] Bonari, J., Kühn, L., von Danwitz, M. & Popp, A. (2024). Towards Real-Time Urban Physics Simulations with Digital Twins, 2024 28th International Symposium on Distributed Simulation and Real Time Applications (DS-RT), pp. 18–25. doi:10.1109/DS-RT62209.2024.00013 [2] Ballarin, F., Sartori, A. & Rozza, G. (2015). RBniCS – reduced order modelling in FEniCS, [Online]. Available: https://www.rbnicsproject.org/ [3] Rozza, G., Ballarin, F., Scandurra, L. & Pichi, F. (2024). Real Time Reduced Order Computational Mechanics, SISSA Springer Series. [4] Demo, N., Tezzele, M. & Rozza, G. (2018). EZyRB: Easy Reduced Basis method (Version 1.1.1). [Computer software]. https://doi.org/10.21105/joss.00661