Multi-Agent Artificial Intelligence for Real-Time Energy Management in Smart Renewable Microgrids

The high penetration rates of renewable energy sources (RES) calls for more advanced and novel‌ technologies for The high penetration rates of renewable energy sources (RES) calls for more advanced and novel technologies for effective management of RES intermittency and variability․ Smart renewable microgrids have emerged as a potential solution for local controllable grids for distributed generation and grid resilience enhancement․ Hence‚ for successful operation‚ smart renewable microgrid systems require effective Real-Time Energy Management (R-TEM) strategies which can account for uncertainties introduced by renewable generation‚ variable load demand and operational constraints․ This paper provides the state-of-the-art review and proposes a Multi-Agent Artificial Intelligence (MA-AI) based R-TEM solution for smart renewable microgrids․ We presents a system architecture of MA-AI‚ showing the profiles and interactions of specific agents‚ such as the Distributed Generator Agent‚ Energy Storage Agent‚ Load Agent‚ and Coordinator Agent․ In this paper‚ we define R-TEM problem with both cost minimizing and emission minimizing objective functions and operational constraints and then propose a MA-AI framework for implementing autonomous and adaptive decision-making based on Multi-Agent Deep Deterministic Policy Gradient(MADDPG) model with Long Short-Term Memory (LSTM) / Transformers forecasting models and improved coordination strategies․ The system was implemented in Python-based simulation environment. The MARL algorithms were implemented using the OpenAI Gym framework‚ alongside the interaction of the code with the energy system models․ Simulations show that this approach improves considerably on rule-based strategies‚ yielding 22.8% cost savings cost savings‚ 17․3% more renewables penetration the grid operate independently and reliably with a 24․8% improvement as well as lowered the total carbon emissions by 18․5% by prioritizing renewable energy․ The paper provides insights on scalability‚ cyber-security‚ explainable AI (XAI)‚ and future work in agentic artificial intelligence‚ quantum AI and hybridizing agents with multi-energy systems (for example‚ electricity and heat)․ This reveals the promise of MA-AI technology for efficient‚ strong‚ and sustainable power management in the emerging ecosystem of smart renewable microgrids.

Authors

Publication Details

Journal
ABUAD Journal of Engineering Research and Development (AJERD)
Published
2026-11-15
DOI
https://doi.org/10.53982/ajerd.2026.0903.25-j
Primary Topic
Microgrid Control and Optimization
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Multi-Agent Artificial Intelligence for Real-Time Energy Management in Smart Renewable Microgrids

Daniel Ogheneovo Johnson, Sunday Oladeji, Oluyemi Petinrin, Joshua Oyebola
ABUAD Journal of Engineering Research and Development (AJERD)
Microgrid Control and Optimization
article

Multi-Agent Artificial Intelligence for Real-Time Energy Management in Smart Renewable Microgrids

Daniel Ogheneovo Johnson, Sunday Oladeji, Oluyemi Petinrin, Joshua Oyebola
article en

Abstract

The high penetration rates of renewable energy sources (RES) calls for more advanced and novel‌ technologies for The high penetration rates of renewable energy sources (RES) calls for more advanced and novel technologies for effective management of RES intermittency and variability․ Smart renewable microgrids have emerged as a potential solution for local controllable grids for distributed generation and grid resilience enhancement․ Hence‚ for successful operation‚ smart renewable microgrid systems require effective Real-Time Energy Management (R-TEM) strategies which can account for uncertainties introduced by renewable generation‚ variable load demand and operational constraints․ This paper provides the state-of-the-art review and proposes a Multi-Agent Artificial Intelligence (MA-AI) based R-TEM solution for smart renewable microgrids․ We presents a system architecture of MA-AI‚ showing the profiles and interactions of specific agents‚ such as the Distributed Generator Agent‚ Energy Storage Agent‚ Load Agent‚ and Coordinator Agent․ In this paper‚ we define R-TEM problem with both cost minimizing and emission minimizing objective functions and operational constraints and then propose a MA-AI framework for implementing autonomous and adaptive decision-making based on Multi-Agent Deep Deterministic Policy Gradient(MADDPG) model with Long Short-Term Memory (LSTM) / Transformers forecasting models and improved coordination strategies․ The system was implemented in Python-based simulation environment. The MARL algorithms were implemented using the OpenAI Gym framework‚ alongside the interaction of the code with the energy system models․ Simulations show that this approach improves considerably on rule-based strategies‚ yielding 22.8% cost savings cost savings‚ 17․3% more renewables penetration the grid operate independently and reliably with a 24․8% improvement as well as lowered the total carbon emissions by 18․5% by prioritizing renewable energy․ The paper provides insights on scalability‚ cyber-security‚ explainable AI (XAI)‚ and future work in agentic artificial intelligence‚ quantum AI and hybridizing agents with multi-energy systems (for example‚ electricity and heat)․ This reveals the promise of MA-AI technology for efficient‚ strong‚ and sustainable power management in the emerging ecosystem of smart renewable microgrids.

ABUAD Journal of Engineering Research and Development (AJERD)Vol. 9(3)
Microgrid Control and Optimization
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon