Modeling renewable energy production and CO2 emissions in the region of Adrar in Algeria using LSTM neural networks - Université Grenoble Alpes
Communication Dans Un Congrès Année : 2023

Modeling renewable energy production and CO2 emissions in the region of Adrar in Algeria using LSTM neural networks

Résumé

Earthquake simulations at the urban scale usually focus on estimating the damages to the built environment and the consequent losses without fully taking into account human behavior in crisis. Yet, human behavior is a key element for improving crisis disaster management; therefore, it is important to include it in seismic crisis simulations. In this study, an agent-based model for the simulation of pedestrian evacuation during earthquakes at the city scale is developed following an interdisciplinary approach. The model recreates the urban conditions using Geographic Information System (GIS) and a synthetic population, in addition to the earthquake consequences on the urban fabric. Moreover, the model integrates realistic human behaviors calibrated using quantitative survey results. We simulate pedestrian outdoor mobility with the different constraints that affect it such as the topography and the presence of debris. The simulator is applied to the case of Beirut, Lebanon. A what-if approach is adopted to analyze the population’s safety in case of earthquakes in Beirut, particularly the open spaces’ capacity to provide shelters and the effect of debris and realistic human behaviors on people’s safety. The simulation results show that less than 40% of the population is able to arrive at an open space within 15 min after an earthquake. This number is further reduced when some open spaces are locked. Debris and realistic human behaviors significantly delay the arrivals to safe areas and, therefore, should not be neglected in earthquake simulations.

Dates et versions

hal-04650795 , version 1 (17-07-2024)

Identifiants

Citer

Seif Eddine Bouziane, Julie Dugdale, Mohamed Tarek Khadir. Modeling renewable energy production and CO2 emissions in the region of Adrar in Algeria using LSTM neural networks. Hawaii International Conference on System Sciences, Jan 2022, Maui (Hawaii), United States. pp.357-377, ⟨10.24251/HICSS.2022.308⟩. ⟨hal-04650795⟩

Collections

UGA
8 Consultations
0 Téléchargements

Altmetric

Partager

More