Quantitative study of drivers and barriers of renewable energy development: Text mining analysis based on web search and crawled samples

Résumé fourni par la source

Accelerating renewable energy deployment is central to climate mitigation and energy security, yet the transition is shaped not only by technology costs but also by evolving institutional, financial, and social constraints. To provide a more systematic account of these multi-dimensional dynamics, this study develops a text-based quantitative framework that combines schema-constrained large-language-model extraction, PESTEL coding, and distributional analysis. We collected 717 candidate web texts and retained 354 highly relevant public records after de-duplication, language filtering, and relevance screening. The final corpus spans 2006–2025 and covers policy documents, industry reports, news, and think-tank analyses. We then mapped extracted evidence on drivers, barriers, stakeholders, metrics, and causal links into a structured analytical database and used intensity scoring, temporal comparison, sectoral composition analysis, co-occurrence mapping, and bootstrap-based long-tail validation to characterize transition patterns. The results show that institutional arrangements and financial instruments are the most consistent drivers of renewable energy deployment, underscoring the centrality of policy design and risk allocation in transition acceleration. Economic factors exhibit a dual role: declining costs support diffusion, while financing barriers, long payback periods, and revenue uncertainty remain major constraints. The time-series results indicate that attention to drivers increased sharply after 2015, whereas barrier-oriented discussion rose later, revealing a clear phase lag between ambition-setting and implementation friction. Sectoral comparison further shows that the power sector is primarily driven by environmental targets and institutional arrangements, while end-use sectors such as transport and industry are more strongly constrained by techno-economic thresholds and infrastructure gaps. In addition, the metric system extracted from the corpus displays a statistically robust long-tail structure, suggesting that social and environmental performance indicators remain under-represented in prevailing evaluation practices. Overall, the study offers both empirical evidence and a reproducible workflow for large-scale text-based assessment of renewable energy transition risks and governance priorities.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.