Enhanced Decision Making in Smart Grid Management by Optimizing Adaptive Multi-Agent Reinforcement Learning with Vehicle-to-Grid Systems

Simple item page
dc.contributor.authorPannee Suanpang
dc.contributor.authorPitchaya Jamjuntr
dc.contributor.correspondenceP. Suanpang; Department of Information Technology, Faculty of Science & Technology, Suan Dusit University, Bangkok, Thailand; email: pannee_sua@dusit.ac.th
dc.date.accessioned2025-03-10T07:34:20Z
dc.date.available2025-03-10T07:34:20Z
dc.date.issued2024
dc.description.abstractThis research proposes a decision-making framework in which the Adaptive Multi-Agent Reinforcement Learning (MARL) model and the concept of Vehicle-to-Grid (V2G) interactivity are employed to improve the effective management of smart grids. The research hypothesis introduces innovations for improving the efficiency and security of power systems in the global south, primarily by controlling the net energy transmission between the defined electric vehicles (EVs) and the grid. Other issues that require attention to ensure the proper functioning of smart grids include demand response, load management, and energy storage optimization. In this instance, these gaps are filled by the systemÕs proposed framework. With the help of MARL, the system dynamics' autonomous learning aspects allow the system to adapt to the capacity of renewable energy sources and electricity demand, which is also time-dependent. Because of the MARL, the autonomous coordination of decision-making has resulted in very positive changes in the system's effectiveness. In particular, this framework permitted an increase of 13.6% in the total energy exchange between EVs and the grid, and the grid stability index improved from 0.84 to 0.87 compared to what would have been achieved with the conventional methods. Enhanced energy management and pricing rehabs added another 22% to net savings. Further, it is stated that deploying MARL-based V2G systems in developing areas has many benefits, including more robust grid reliability and energy security and better integration of renewable energy resources. Such changes aid in reducing fossil fuel use and greenhouse gas emissions. © 2024 Regional Association for Security and crisis management. All rights reserved.
dc.identifier.citationDecision Making: Applications in Management and Engineering
dc.identifier.doi10.31181/dmame7120241257
dc.identifier.issn25606018
dc.identifier.scopus2-s2.0-85214667475
dc.identifier.urihttps://repository.dusit.ac.th//handle/123456789/4472
dc.languageEnglish
dc.publisherRegional Association for Security and crisis management
dc.rights.holderScopus
dc.subjectDecision making
dc.subjectOptimization, Multi-Agent Reinforcement Learning
dc.subjectSmart grid management
dc.subjectVehicle-to-Grid Systems
dc.titleEnhanced Decision Making in Smart Grid Management by Optimizing Adaptive Multi-Agent Reinforcement Learning with Vehicle-to-Grid Systems
dc.typeArticle
mods.location.urlhttps://www.scopus.com/inward/record.uri?eid=2-s2.0-85214667475&doi=10.31181%2fdmame7120241257&partnerID=40&md5=21777500a433ed38700bb1e30d18932e
oaire.citation.endPage530
oaire.citation.issue1
oaire.citation.startPage494
oaire.citation.volume7
Files
Collections
SCOPUS 2024