Zhao, S. orcid.org/0000-0001-9660-6821, Li, K., Yu, J. et al. (1 more author) (2024) Scheduling of futuristic railway microgrids—A FRA-pruned twins-actor DDPG approach. Energy, 313. 134089. ISSN 0360-5442
Abstract
Transport decarbonization including railway electrification has become a global trend in tackling the climate change challenge due to the substantive green-house-gas emissions from the transport sector. However, within the context of the rapidly growing electricity demand due to railway decarbonization, the existing power supply infrastructures are often stretched beyond their rating capacity, thus slowing down the railway electrification pace. To address this challenge, this paper presents a novel railway microgrid solution to economically and efficiently cater for the growing electricity demands of the railway sector. The proposed railway microgrid uses the existing traction network, local renewable generations and local power grid as its energy sources to meet both the traction and non-traction loads in the electrified railway systems. It also employs hydrogen storage systems and aggregated electric vehicle (EV) fleets to support flexible energy scheduling. To address the uncertainty of renewable energy and implement efficient energy scheduling, machine learning tools and agents are introduced in formulating and solving the optimal scheduling problem, including bidirectional long short-term memory (BiLSTM) and an improved twins-actor deep deterministic policy gradient (Twins-Actor DDPG) method. To reduce the overall computational complexity, a fast recursive algorithm (FRA) is adopted to streamline the machine learning agents. Case studies confirm that the proposed optimal microgrid operation can achieve up to 18.7% costs reduction for the daily operation of the railway power supply system while meeting the electricity demand, and FRA can achieve up to 78.6% agent size reduction and 93.1% computational cost savings.
Metadata
Item Type: | Article |
---|---|
Authors/Creators: |
|
Copyright, Publisher and Additional Information: | This is an author produced version of an article published in Energy, made available under the terms of the Creative Commons Attribution License (CC-BY), which permits unrestricted use, distribution and reproduction in any medium, provided the original work is properly cited. |
Keywords: | Railway power supply systems, Microgrid, Machine learning, Renewable generation, Energy storage |
Dates: |
|
Institution: | The University of Leeds |
Academic Units: | The University of Leeds > Faculty of Engineering & Physical Sciences (Leeds) > School of Electronic & Electrical Engineering (Leeds) |
Depositing User: | Symplectic Publications |
Date Deposited: | 10 Dec 2024 12:27 |
Last Modified: | 20 Feb 2025 11:15 |
Status: | Published |
Publisher: | Elsevier BV |
Identification Number: | 10.1016/j.energy.2024.134089 |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:220658 |