2025-09-11
2026-06-18
2025-12-04
Manuscript received August 6, 2026; accepted September 16, 2026; published November 20, 2026
Abstract—This systematic review synthesizes evidence on multi-agent artificial intelligence for autonomous smart-grid operation and renewable-energy coordination, covering publications from 1 January 2023 to 31 August 2026. Following PRISMA 2020 and PRISMA-S, 39 studies were included in the qualitative synthesis. The review distinguishes traditional Multi-agent Systems (MAS), Multi-agent Reinforcement Learning (MARL), generative AI-augmented MAS, and emerging Agentic AI. Peer-reviewed evidence supports applications in voltage and frequency regulation, distributed energy management, demand response, electric-vehicle coordination, and service restoration, although most evaluations remain simulation-based and heterogeneous. Preprints are explicitly identified and are used mainly for rapidly evolving Agentic AI, LLM-agent, communication, and synthetic-data topics. Excluding preprints does not materially change the conclusions for established MAS/MARL applications, but it substantially weakens claims of deploymentready Agentic AI. The most credible deployment pathway is therefore a layered architecture combining deterministic safety constraints, distributed learning, validated engineering solvers, cybersecurity controls, and human oversight. Regulatory compliance, market rules, lifecycle economics, workforce readiness, data governance, standardized benchmarks, and field validation remain decisive barriers to scalable adoption. Keywords—multi-agent systems, agentic AI, smart grids, multi-agent reinforcement learning, renewable energy coordination Cite: Kwadwo Larbi, "Multi-agent AI Systems for Autonomous Smart Grid Operation and Renewable Energy Coordination," International Journal of Smart Grid and Clean Energy, Vol. 15, No. 2, pp. 123-145, 2026. doi: 10.12720/sgce.15.2.123-145 Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).