Evolution of Hybrid Renewable Energy Systems: A Maturity–Momentum Synthesis of Technologies, Storage, Optimization, Artificial Intelligence and Digitalization

Hybrid renewable energy systems (HRES) now sit at the centre of efforts to decarbonize electricity supply. The literature describing them has grown faster than anyone has organized it. Work on storage integration, optimization, artificial intelligence (AI) and digital energy management has accumulated largely in parallel, and existing reviews tend to treat each strand on its own. This paper takes on that fragmentation through a structured, bibliometrically grounded synthesis instead of new experimental work. What it offers is a maturity–momentum framework that places every research theme on two axes at once: accumulated citation influence and current publication growth. The position a theme occupies separates settled knowledge from frontiers that are busy but still empirically thin. The framework was applied to a corpus of 18,420 Scopus-indexed documents published between 2000 and 2024, retrieved through eight pre-specified TITLE-ABS-KEY queries and screened against documented eligibility criteria. From that corpus, 150 studies were read in full and synthesized thematically. The corpus spans 1840 source journals and 142 countries, carries 312,640 citations, yields a field-level h-index of 214, and grew at a compound annual rate of 18.3% between 2015 and 2024. One relationship runs through the whole analysis: citation maturity and publication momentum move in opposite directions. Early HRES design and pumped hydro storage record the highest citation intensity, at 48.2 and 42.1 citations per document, while their output decelerates. AI-driven operation and future-oriented themes grow fastest, with AI-related output rising roughly 65-fold between 2010 and 2024, yet they rest on comparatively thin citation foundations. Reported minimum levelized cost of energy across successive families of optimization methods fell from USD 0.091 kWh−1 in 2019 to USD 0.037 kWh−1 in 2024, and the reductions arrived in steps that coincide with changes in the method family, not with incremental algorithmic refinement. Eight gaps persist: large-scale experimental validation, multi-energy interoperability, uncertainty-aware optimization, cybersecurity of AI-enabled control, Digital Twin standardization, explainable AI, lifecycle sustainability assessment, and integrated electricity–hydrogen market design. Taken together, they indicate that the weakest point in the field is no longer algorithmic capability. It is the shortage of long-horizon field evidence against which simulated gains can be tested.