A Graph-Based Mobility Model for Electric Vehicles in Urban Traffic Networks: Application to the Grenoble Metropolitan Area

Rodriguez-Vega, Martin; Canudas-de-Wit, Carlos; De Nunzio, Giovanni; Othman, Bassel · 2023 · Crossref

DOI: 10.23919/ecc57647.2023.10178345

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper introduces a graph-based electromobility model designed to simulate the movement and State-of-Charge (SoC) evolution of electric vehicles (EVs) within urban traffic networks. The research addresses the need for accurate spatiotemporal modeling of EV energy demand to support infrastructure planning and Vehicle-to-Grid (V2G) integration. Existing models often treat mobility and energy consumption separately; this work couples human mobility patterns with EV energy dynamics to predict where and when energy demand occurs. The model is applied to the Grenoble Metropolitan Area in France to evaluate the capacity of the current charging infrastructure under increasing EV penetration rates. The methodology employs a dynamic graph formulation consisting of seven interconnected modules. The model tracks the number of EVs, flow between nodes, and aggregated SoC across a network of origin (residential) and destination (work, school, leisure, etc.) nodes. It integrates a people mobility module based on conservation Ordinary Differential Equations, a mode choice module using a logit model to estimate private car usage, and an EV mobility module that converts human flows into vehicle flows based on penetration rates. Energy consumption is calculated using a power loss module that accounts for vehicle mass, aerodynamics, road inclination, and auxiliary systems. Charging inputs are modeled for home, office, and public stations, with public charger allocation determined by a gravity attraction law based on distance and power capacity. The model was calibrated using multi-source public data, including census statistics, household mobility surveys, OpenStreetMap road networks, and charger location databases. Experiments were conducted via simulation to determine the maximum sustainable EV penetration rate for the Grenoble area. The study identified a "phase transition" point where the existing charging infrastructure can no longer meet the energy demand of the EV fleet. Results indicate that the current infrastructure supports a specific penetration threshold, beyond which the average SoC of the network drops significantly, indicating insufficient charging capacity. The simulations revealed spatial and temporal patterns of energy distribution, showing concentration in central areas during working hours and redistribution to residential zones at night. The analysis highlighted that while some peripheral municipalities show high penetration rates, the central city of Grenoble, despite having a lower penetration rate, holds the majority of EVs and thus drives the primary energy demand. The significance of this work lies in providing a comprehensive tool for urban planners and grid operators to assess the scalability of EV adoption. By linking human mobility with energy consumption, the model offers a more accurate prediction of grid stress points than models considering only vehicle counts or only traffic flow. The findings demonstrate the limitations of the current charging infrastructure in Grenoble, providing a quantitative basis for future infrastructure expansion and V2G strategy development. This approach can be generalized to other metropolitan areas to optimize the transition toward carbon-neutral transportation networks.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• v2x connected vehicle