EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A lightweight digital twin for energy optimization and predictive maintenance in urban rail networks using AI, IoT, and blockchain

Rita Belkadi, Atmane El Hadb, Mohammed Hatim Rziki, Moulay Brahim Sedra, Mohamed Khalifa Boutahir

Clean Energy Science and Technology · 05.11.2026

DOI: 10.18686/cest842 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Clean Energy Science and Technology, online erschienen am 05.11.2026. Verfasst von Rita Belkadi und 4 weiteren Autor:innen. Die Arbeit nennt 26 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the rapid urbanization of modern cities, public transport demand has significantly increased, leading to higher energy consumption and accelerated equipment degradation, affecting system efficiency and sustainability. In fact, urban rail networks generally rely on static models and centralized systems unable to deal with real-time fluctuations. This paper proposes a lightweight digital twin (DT) framework for urban rail transit systems, integrating artificial intelligence (AI), the Internet of Things (IoT), and blockchain technology to enable real-time energy optimization and predictive maintenance. IoT sensors continuously collect operational and energy data across the network. Machine learning models, including Random Forests and Neural Networks, dynamically predict energy demand, while a novel genetic algorithm (GA) optimizes train synchronization to maximize regenerative braking energy recovery. Moreover, a blockchain layer is integrated to ensure the security of critical energy transactions and maintenance events. The approach is validated through DT simulations using 2.7 million operational records from the Paris metropolitan rail (Metro) systems (Régie Autonome des Transports Parisiens/Paris Transport Authority, RATP). Simulation results showed an increase in regenerative energy utilization by 92.4%, a reduction in overall peak-hour energy consumption by 12.7%, and an improvement in the early detection of mechanical anomalies by 24% compared to traditional baseline models. The integration of a lightweight DT combining AI, IoT, and blockchain has the potential to provide a more scalable, more resilient, and overall, more economically effective system for urban rail networks. By offering an adaptive, data-driven digital representation, the proposed framework supports sustainable transportation initiatives and can be generalized to other global urban railway networks.

Zitieren

APA

Belkadi, R., Hadb, A. E., Rziki, M. H., Sedra, M. B., & Boutahir, M. K. (2026). A lightweight digital twin for energy optimization and predictive maintenance in urban rail networks using AI, IoT, and blockchain. Clean Energy Science and Technology. https://doi.org/10.18686/cest842

Vancouver

Belkadi R, Hadb AE, Rziki MH, Sedra MB, Boutahir MK. A lightweight digital twin for energy optimization and predictive maintenance in urban rail networks using AI, IoT, and blockchain. Clean Energy Science and Technology. 2026. doi:10.18686/cest842

BibTeX
@article{Belkadi2026,
  author = {Belkadi, Rita and Hadb, Atmane El and Rziki, Mohammed Hatim and Sedra, Moulay Brahim and Boutahir, Mohamed Khalifa},
  title = {A lightweight digital twin for energy optimization and predictive maintenance in urban rail networks using AI, IoT, and blockchain},
  journal = {Clean Energy Science and Technology},
  year = {2026},
  doi = {10.18686/cest842}
}
Kostenlos starten →