Please use this identifier to cite or link to this item: http://bura.brunel.ac.uk/handle/2438/24944
Title: Profit maximization for large-scale energy storage systems to enable fast EV charging infrastructure in distribution networks
Authors: Lai, CS
Chen, D
Zhang, J
Zhang, X
Xu, X
Taylor, G
Lai, LL
Keywords: distribution network optimization;fast EV charging demand;deep reinforcement learning;battery energy storage systems
Issue Date: 5-Aug-2022
Publisher: Elsevier
Citation: Lai, C.S., Chen, D., Zhang, J., Zhang, X., Xu, X., Taylor, G. and Lai, L.L. (2022) 'Profit maximization for large-scale energy storage systems to enable fast EV charging infrastructure in distribution networks', Energy, 0 (in press), 124852, pp. 1-64. doi: 10.1016/j.energy.2022.124852.
Abstract: Coppyright © 2022 The Author(s). Large-scale integration of battery energy storage systems (BESS) in distribution networks has the potential to enhance the utilization of photovoltaic (PV) power generation and mitigate the negative effects caused by electric vehicles (EV) fast charging behavior. This paper presents a novel deep reinforcement learning-based power scheduling strategy for BESS which is installed in an active distribution network. The network includes fast EV charging demand, PV power generation, and electricity arbitrage from main grid. The aim is to maximize the profit of BESS operator whilst maintaining voltage limits. The novel strategy adopts a Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm and requires forecasted PV power generation and EV smart charging demand. The proposed strategy is compared with Deep Deterministic Policy Gradient (DDPG), Particle Swarm Optimization and Simulated Annealing algorithms to verify its effectiveness. Case studies are conducted with smart EV charging dataset from Project Shift (UK Power Networks Innovation) and the UK photovoltaic dataset. The Internal Rate of Return results with TD3 and DDPG algorithms are 9.46% and 8.69%, respectively, which show that the proposed strategy can enhance power scheduling and outperforms the mainstream methods in terms of reduced levelized cost of storage and increased net present value.
URI: https://bura.brunel.ac.uk/handle/2438/24944
DOI: https://doi.org/10.1016/j.energy.2022.124852
ISSN: 0360-5442
Other Identifiers: 124852
Appears in Collections:Brunel OA Publishing Fund
Dept of Electronic and Computer Engineering Research Papers

Files in This Item:
File Description SizeFormat 
FullText.pdf2.68 MBAdobe PDFView/Open


This item is licensed under a Creative Commons License Creative Commons