Next Article in Journal
Data-Driven ANN Model Development for Maximum Power Point Estimation in PV Panel Under Partial Shading Conditions
Previous Article in Journal
Physics-Informed Deep Reinforcement Learning for Compact VBT Farms: Integration, Power Quality, and Economics
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Engineering Proceedings
Volume 144
Issue 1
10.3390/engproc2026144009
engproc-logo
Submit to this Journal
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Proceeding Paper

Physics-Constrained Multi-Agent Deep Reinforcement Learning for Real-Time Energy Management of a Saharan Hybrid Microgrid †

by
Redouane Mihramane
*,
S. Salah Ech-Charqaouy
,
Abdelkader Boulezhar
,
Amjad Ech-Charqaouy
and
Nizar Ech-Charqaouy
Faculty of Sciences Ain Chock, Hassan II University, Casablanca 20100, Morocco
*
Author to whom correspondence should be addressed.
†
Presented at the 2nd International Conference on Sciences and Techniques for Renewable Energy and the Environment, Al Hoceima, Morocco, 28–30 April 2026.
Eng. Proc. 2026, 144(1), 9; https://doi.org/10.3390/engproc2026144009
Published: 25 June 2026
(This article belongs to the Proceedings of 2nd International Conference on Sciences and Techniques for Renewable Energy and the Environment)
Browse Figures
Versions Notes

Abstract

This paper addresses the challenge of ensuring physically feasible and reliable real-time control of hybrid microgrids in harsh desert environments. A physics-constrained multi-agent Deep Q-Network (MA-DQN) is proposed for energy management of a grid-interactive microgrid in the Moroccan Sahara. The method embeds operational constraints directly into learning through action filtering, penalty-aware rewards, and coordinated PCC control. The results show a reduction in operational cost from 1250 MAD to 1120 MAD (−10.4%) and CO2 emissions from 318.9 kg to 272.5 kg (−14.6%), while maintaining voltage within ±10% limits and eliminating PCC oscillations. The framework delivers stable, reliable, and deployment-ready control.
Keywords: hybrid microgrid; deep reinforcement learning; multi-agent systems; physics-constrained control; energy management system; renewable energy integration hybrid microgrid; deep reinforcement learning; multi-agent systems; physics-constrained control; energy management system; renewable energy integration

Share and Cite

MDPI and ACS Style

Mihramane, R.; Ech-Charqaouy, S.S.; Boulezhar, A.; Ech-Charqaouy, A.; Ech-Charqaouy, N. Physics-Constrained Multi-Agent Deep Reinforcement Learning for Real-Time Energy Management of a Saharan Hybrid Microgrid. Eng. Proc. 2026, 144, 9. https://doi.org/10.3390/engproc2026144009

AMA Style

Mihramane R, Ech-Charqaouy SS, Boulezhar A, Ech-Charqaouy A, Ech-Charqaouy N. Physics-Constrained Multi-Agent Deep Reinforcement Learning for Real-Time Energy Management of a Saharan Hybrid Microgrid. Engineering Proceedings. 2026; 144(1):9. https://doi.org/10.3390/engproc2026144009

Chicago/Turabian Style

Mihramane, Redouane, S. Salah Ech-Charqaouy, Abdelkader Boulezhar, Amjad Ech-Charqaouy, and Nizar Ech-Charqaouy. 2026. "Physics-Constrained Multi-Agent Deep Reinforcement Learning for Real-Time Energy Management of a Saharan Hybrid Microgrid" Engineering Proceedings 144, no. 1: 9. https://doi.org/10.3390/engproc2026144009

APA Style

Mihramane, R., Ech-Charqaouy, S. S., Boulezhar, A., Ech-Charqaouy, A., & Ech-Charqaouy, N. (2026). Physics-Constrained Multi-Agent Deep Reinforcement Learning for Real-Time Energy Management of a Saharan Hybrid Microgrid. Engineering Proceedings, 144(1), 9. https://doi.org/10.3390/engproc2026144009

Article Metrics

Back to TopTop