Entropy-driven multi agent deep reinforcement learning for resilient distribution networks: coordinating MESS and microgrids

In extreme disasters where severe main grid failures lead to widespread power outages in distribution networks, rapid critical load restoration (CLR) becomes crucial for enhancing power supply reliability. Aiming to improve distribution network resilience, this paper proposes an entropy-driven multi...

Full description

Saved in:
Bibliographic Details
Main Authors: Changchun Cai, Fenglu Gan, Yong Cui, Bo Li, Shixi Hou
Format: Article
Language:English
Published: Elsevier 2025-09-01
Series:International Journal of Electrical Power & Energy Systems
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S0142061525005162
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849407372090408960
author Changchun Cai
Fenglu Gan
Yong Cui
Bo Li
Shixi Hou
author_facet Changchun Cai
Fenglu Gan
Yong Cui
Bo Li
Shixi Hou
author_sort Changchun Cai
collection DOAJ
description In extreme disasters where severe main grid failures lead to widespread power outages in distribution networks, rapid critical load restoration (CLR) becomes crucial for enhancing power supply reliability. Aiming to improve distribution network resilience, this paper proposes an entropy-driven multi-agent deep reinforcement learning (MADRL) framework coordinating mobile energy storage systems (MESS) and microgrid reconfiguration. First, with critical load restoration as the objective function, coordinated optimization model for MESS dispatch and network reconfiguration is constructed, which comprehensively considers security constraints of both distribution networks and microgrids. Subsequently, the coordinated optimization problem is formulated as a Markov Decision Process (MDP). Then, a multi-agent deep Q-learning (MADQL) algorithm is developed to search for optimal strategies, featuring a topology aware entropy-driven exploration (TAEE) mechanism to discover high-value actions and accelerate training convergence. Additionally, an action masking technique is introduced to enforce operational safety by dynamically filtering constraint-violating actions. Finally, extensive numerical results validate the effectiveness of our proposed method.
format Article
id doaj-art-979b260c08804e43b9cd4fa2ceec4361
institution Kabale University
issn 0142-0615
language English
publishDate 2025-09-01
publisher Elsevier
record_format Article
series International Journal of Electrical Power & Energy Systems
spelling doaj-art-979b260c08804e43b9cd4fa2ceec43612025-08-20T03:36:06ZengElsevierInternational Journal of Electrical Power & Energy Systems0142-06152025-09-0117011096810.1016/j.ijepes.2025.110968Entropy-driven multi agent deep reinforcement learning for resilient distribution networks: coordinating MESS and microgridsChangchun Cai0Fenglu Gan1Yong Cui2Bo Li3Shixi Hou4College of Artificial Intelligence and Automation, Hohai University, Jiangsu, China; Jiangsu Key Laboratory of Power Transmission & Distribution Equipment Technology, Hohai University, Jiangsu, China; Corresponding author.College of Artificial Intelligence and Automation, Hohai University, Jiangsu, China; Jiangsu Key Laboratory of Power Transmission & Distribution Equipment Technology, Hohai University, Jiangsu, ChinaCollege of Management, Anhui Science and Technology University, Anhui, ChinaCollege of Artificial Intelligence and Automation, Hohai University, Jiangsu, China; Jiangsu Key Laboratory of Power Transmission & Distribution Equipment Technology, Hohai University, Jiangsu, ChinaCollege of Artificial Intelligence and Automation, Hohai University, Jiangsu, China; Jiangsu Key Laboratory of Power Transmission & Distribution Equipment Technology, Hohai University, Jiangsu, ChinaIn extreme disasters where severe main grid failures lead to widespread power outages in distribution networks, rapid critical load restoration (CLR) becomes crucial for enhancing power supply reliability. Aiming to improve distribution network resilience, this paper proposes an entropy-driven multi-agent deep reinforcement learning (MADRL) framework coordinating mobile energy storage systems (MESS) and microgrid reconfiguration. First, with critical load restoration as the objective function, coordinated optimization model for MESS dispatch and network reconfiguration is constructed, which comprehensively considers security constraints of both distribution networks and microgrids. Subsequently, the coordinated optimization problem is formulated as a Markov Decision Process (MDP). Then, a multi-agent deep Q-learning (MADQL) algorithm is developed to search for optimal strategies, featuring a topology aware entropy-driven exploration (TAEE) mechanism to discover high-value actions and accelerate training convergence. Additionally, an action masking technique is introduced to enforce operational safety by dynamically filtering constraint-violating actions. Finally, extensive numerical results validate the effectiveness of our proposed method.http://www.sciencedirect.com/science/article/pii/S0142061525005162Distribution systemNature disasterDeep reinforcement learningMicrogridsMobile energy storageResilience
spellingShingle Changchun Cai
Fenglu Gan
Yong Cui
Bo Li
Shixi Hou
Entropy-driven multi agent deep reinforcement learning for resilient distribution networks: coordinating MESS and microgrids
International Journal of Electrical Power & Energy Systems
Distribution system
Nature disaster
Deep reinforcement learning
Microgrids
Mobile energy storage
Resilience
title Entropy-driven multi agent deep reinforcement learning for resilient distribution networks: coordinating MESS and microgrids
title_full Entropy-driven multi agent deep reinforcement learning for resilient distribution networks: coordinating MESS and microgrids
title_fullStr Entropy-driven multi agent deep reinforcement learning for resilient distribution networks: coordinating MESS and microgrids
title_full_unstemmed Entropy-driven multi agent deep reinforcement learning for resilient distribution networks: coordinating MESS and microgrids
title_short Entropy-driven multi agent deep reinforcement learning for resilient distribution networks: coordinating MESS and microgrids
title_sort entropy driven multi agent deep reinforcement learning for resilient distribution networks coordinating mess and microgrids
topic Distribution system
Nature disaster
Deep reinforcement learning
Microgrids
Mobile energy storage
Resilience
url http://www.sciencedirect.com/science/article/pii/S0142061525005162
work_keys_str_mv AT changchuncai entropydrivenmultiagentdeepreinforcementlearningforresilientdistributionnetworkscoordinatingmessandmicrogrids
AT fenglugan entropydrivenmultiagentdeepreinforcementlearningforresilientdistributionnetworkscoordinatingmessandmicrogrids
AT yongcui entropydrivenmultiagentdeepreinforcementlearningforresilientdistributionnetworkscoordinatingmessandmicrogrids
AT boli entropydrivenmultiagentdeepreinforcementlearningforresilientdistributionnetworkscoordinatingmessandmicrogrids
AT shixihou entropydrivenmultiagentdeepreinforcementlearningforresilientdistributionnetworkscoordinatingmessandmicrogrids