DTE AICCOMAS 2025

Numerical Strategies for Cities and Territories using Digital Twin: from Urban Planning to Enhanced management

  • Waeytens, Julien (Université Gustave Eiffel)
  • Hamada, Tsubasa (Université Gustave Eiffel)
  • Chakir, Rachida (Université Gustave Eiffel)
  • Perrin, Guillaume (Université Gustave Eiffel)
  • Lévèque, Philippe (Université Gustave Eiffel)
  • Bourgeois, Emmanuel (Université Gustave Eiffel)
  • Siegert, Dominique (Université Gustave Eiffel)

Please login to view abstract download link

The digital twin is an expanding tool for cities and territories. It can play a central role to tackle the multidisciplinary and multiscale issues: energy in districts and buildings, outdoor/indoor air quality, drinking and waste water, etc. The integration of sensor data and physical modelling in urban digital twin makes it possible (i) to evaluate an existing urban planning, (ii) to test virtually different urban planning scenarios and (iii) to facilitate the real-time management of city and territories components. Two applications on cities/territories and associated numerical strategies will be presented. Firstly, the digital twin of a Paris district is considered for a detailed and realistic cartography of traffic pollutant and for a smart placement of ZnO photocatalytic panels for a goal-oriented mitigation of air pollution [1]. The proposed numerical strategy is based on 3D turbulent fluid mechanics simulations and sensitivity analysis through adjoint framework. In this study, we observe that depolluting surfaces should be placed on the road, on certain sidewalks of streets (according to street canyon airflows) and on lower parts of building façade in order to reduce human exposure to traffic pollutant in the street and in the apartment/office. Secondly, the digital twin of a light instrumented masonry bridge is studied. In fact, French territories have to manage a large number of aging bridges with a very limited budget. Hence, we propose to solve an optimization problem using the numerical mock-up of the bridge and solid mechanics modelling to select a proper placement of a restricted number of strain sensors in view of detecting a maximum of listed damaged scenarios. Then, after sensor deployment, damage evolutions are monitored through real-time model updating of parameters, e.g. Young Modulus. This approach is illustrated on a real instrumented masonry bridge in Gagnac, southwest France region.