AI–Optimization Coupling in Hybrid Renewable Energy Systems: A Critical Review and Design Framework for Uncertainty, Constraints, and Risk-Aware Scheduling
Year: 2026
Type: article
Abstract: Hybrid renewable energy systems (HRESs) increasingly combine artificial intelligence (AI) with mathematical optimization, but existing surveys often foreground algorithm families rather than the decision architecture by which learning influences optimization and executed action. This critical integrative review proposes a compositional description combining functional motif × coupling direction × coupling stack that integrates five dominant learned-role motifs (P1–P5) and two cross-cutting modifiers, downstream feasibility authority (P6) and physics-/constraint-informed learning (P7). The Scopus-centered evidence assembly frozen on 20 September 2026 contains 1154 unique records; a 50-record pilot informed a purposive 120-record priority corpus, of which 82 records were assessed for boundary and framework testing (49 for Priority A; 33 for Priority B), including negative, anchor and provisional records. The record-level audit identifies 52 positive non-provisional cases but heterogeneous evidence depth; the subset with clearly documented full-text evidence is too limited to independently confirm framework-wide stability. The synthesis maps how AI contributes through better decision inputs, model surrogates, solver guidance, optimization approximation or learned action selection and proposes conditional architecture selection recommendations with explicit evidence and validation obligations. No field prevalence, causal algorithm superiority or full-literature saturation claim is made.
Source: Energies
DOI: 10.3390/en19225538
Authors: Tymoteusz Igor Miller, Irmina Durlik
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $3,137
OpenAlex ID: W7235608943
