Soutenance de Thèse, Mhamad SHMAYSANI, 15 Déc. 2025

Développement d’une supervision énergétique d’un smart-grid ferroviaire à courant continu – RACCOR-D

lundi 15 décembre 2025 à 9h30
JUNIA Bâtiment Colson 1 16 Rue Colson, 59800, Lille
Amphithéâtre Mathilda

Mots-clés :

Réseau smart-grid, Supervision énergétique, Gestion de l’énergie, Réseaux DC, Système électrique ferroviaire, Optimisation énergétique

Résumé :

As railway traffic and energy costs continue to rise, SNCF Réseau wants to improve the performance of its traction substations while meeting European CO₂ reduction targets. Although the main railway lines use AC electrification, a large part of the French network still operates with DC systems. The latter is characterized by relatively low voltage and limited traffic density. The RACCOR-D project aims to address these issues by preparing the move toward higher-voltage DC networks (above 1500 V). As part of this effort, SNCF Réseau plans to upgrade its substations and turn them into smart, modern facilities. These new substations will include photovoltaic (PV) panels and energy storage systems (ESS). Their goal is to reduce dependence on the national electricity grid, lower operating costs, and avoid financial penalties for exceeding contracted power limits. Integrating both ESS and PV into a railway substation creates several challenges. To fully benefit from these technologies, a smart energy management system is essential. It must operate the ESS in a way that maximizes the use of local PV production while helping SNCF Réseau achieve its goals of reducing energy consumption and avoiding penalty charges. In this context, the objective of this PhD thesis is to design an energy supervision system specifically for DC railway networks. The goal is to optimize how the ESS is used in two RACCOR-D possible architectures, where the PV system and the storage are connected directly to the DC bus of the substation and when connected between two traction substations. The supervisor aims to improve both the energy performance and the operating costs of the traction substation, while respecting the technical constraints of DC systems. The supervision strategy is developed with two temporal levels, the long term (day ahead) and the short term (real time). The supervision framework was validated through real-time simulations using OPAL-RT. And to support industrial deployment, a SCADA-based architecture was designed to ensure reliable communication with all smart grid components. This industrial setup enables the supervisor to operate on an industrial PC and to be integrated into the railway substation demonstrator, paving the way for large-scale implementation of intelligent DC substations.