Hazardous atmospheric dispersion in urban areas: A Deep Learning approach for emergency pollution forecast. (June 2022)

• Record Type: Journal Article
• Authors: Mendil, Mouhcine; Leirens, Sylvain; Armand, Patrick; Duchenne, Christophe
• Is Part Of: Environmental modelling & software. Volume 152(2022)
• Publication Date: 2022-06
• Subjects: Deep learning -- Hazardous pollutant -- Dispersion simulation -- Surrogate model -- Synthetic data; ADE Advection-Diffusion equation -- AI Artificial Intelligence -- ANN Artificial Neural Network -- CFD Computational Fluid Dynamics -- DL Deep Learning -- DNN Deep Neural Network -- ML Machine Learning -- MSE Mean Squared Error -- PMSS Parallel Micro-SWIFT-SPRAY; Environmental monitoring -- Computer programs -- Periodicals; Ecology -- Computer simulation -- Periodicals; Digital computer simulation -- Periodicals; Computer software -- Periodicals; Environmental Monitoring -- Periodicals; Computer Simulation -- Periodicals; Environnement -- Surveillance -- Logiciels -- Périodiques; Écologie -- Simulation, Méthodes de -- Périodiques; Simulation par ordinateur -- Périodiques; Logiciels -- Périodiques; Computer software; Digital computer simulation; Ecology -- Computer simulation; Environmental monitoring -- Computer programs; Periodicals; Electronic journals; 363.70015118
• DOI: 10.1016/j.envsoft.2022.105387
• Languages: English
• ISSNs: 1364-8152

Abstract

Abstract: Today, Computational Fluid Dynamics approaches have a high level of spatial/temporal accuracy in modelling atmospheric transport and dispersion in very complex environments. Several numerical models require, however, heavy computational resources and prolonged simulation time up to several days. This time constraint is specifically crucial for intervention planning in case of accidental or malevolent toxic releases in a city. In this paper, we propose to use synthetic data generated by a realistic 3-D transport/dispersion simulator, to train a learning framework called MCxM . The latter relies on a sequence of masking and correction operations to progressively apply the spatial constraints and underlying physics of transport and dispersion. The learning phase uses the urban geometry of the French city Grenoble. We then test the effectiveness of the trained MCxM in a different French city: Paris. The results show that the MCxM 's forecasts are virtually instantaneous and generalize successfully to unseen conditions. Graphical abstract: Image 1 Highlights: Our data-driven surrogate model estimates air pollution exposure in various urban areas. Model construction relies on a physics-informed deep-learning framework. Model training and validation use realistic 3-D synthetic data. Model's speed and accuracy enables emergency planning in case of urban toxic release.