Renewable microgrid optimization using AI: A B-SLR approach and future research directions

Document Type

Article

Publication Title

Energy Reports

Abstract

This study advances the discourse on artificial intelligence (AI) in renewable microgrid optimization by providing a business-centric and critically integrative review rather than a purely technical summary. Existing literature remains fragmented across engineering and policy domains, often overlooking the economic and governance dimensions shaping AI adoption. Employing a Bibliometric–Systematic Literature Review (B-SLR) of 59 peer-reviewed studies published between 2014 and 2025; this paper combines bibliometric mapping with thematic synthesis to assess not only how AI optimizes microgrids but also why institutional and contextual factors influence its real-world effectiveness. Findings reveal a clear shift from deterministic optimization models to hybrid, multi-objective frameworks integrating machine learning, evolutionary algorithms, and game-theoretic approaches. Yet, these advances remain constrained by limited field validation, weak scalability, and insufficient attention to explainability, ethics, and regulatory adaptation, particularly in low- and middle-income contexts. The review contributes an integrated conceptual framework linking AI technique, operational performance, business value, and sustainability outcomes, and proposes a forward-looking agenda emphasizing explainable AI, federated learning, edge computing, and participatory governance. By reframing AI-enabled microgrids as strategic instruments of sustainable modernization, the study highlights how technological innovation, economic feasibility, and ethical governance must converge to achieve inclusive and resilient energy transitions.

First Page

4963

Last Page

4975

DOI

10.1016/j.egyr.2025.11.087

Publication Date

12-1-2025

This document is currently not available here.

Share

COinS