Analyses of Green Energy Innovation Using Patent Data

Analyses of Green Energy Innovation Using Patent Data

ES评分 0

DOI 10.20900/jsr20260029
刊名
JSR
年,卷(期) 2026, 8(2)
作者
作者单位

School of Business, Singapore University of Social Sciences, Singapore 599494, Singapore

摘要
Global reliance on fossil fuels continues to drive greenhouse gas emissions and climate change, necessitating accelerated innovation in green energy technologies to achieve sustainable development and net-zero targets. Despite extensive research on sustainability, systematically integrating patent intelligence with semantic analytics to understand technological evolution in this domain has received limited attention. This study addresses this gap by developing a Multi-Layer Patent Intelligence and Semantic Analytics (MPISA) framework, integrating patent analysis and large language model (LLM)-based semantic clustering using SciBERT, to examine green energy innovation patterns based on patent data from the United States Patent and Trademark Office from 2005 to 2025. Findings reveal a technological reorientation from diversified renewables toward photovoltaic consolidation and subsequent material-driven optimisation. Assignee analysis demonstrates shifting competitive dynamics, with traditional manufacturers declining as power electronics specialists gain prominence. LLM-based thematic clustering indicates that green energy innovation is structured as a hierarchical technology stack rather than a collection of isolated fields. The MPISA framework offers a replicable methodology for understanding technological evolution and informs strategic decision-making for advancing sustainable energy transitions.
Abstract
Global reliance on fossil fuels continues to drive greenhouse gas emissions and climate change, necessitating accelerated innovation in green energy technologies to achieve sustainable development and net-zero targets. Despite extensive research on sustainability, systematically integrating patent intelligence with semantic analytics to understand technological evolution in this domain has received limited attention. This study addresses this gap by developing a Multi-Layer Patent Intelligence and Semantic Analytics (MPISA) framework, integrating patent analysis and large language model (LLM)-based semantic clustering using SciBERT, to examine green energy innovation patterns based on patent data from the United States Patent and Trademark Office from 2005 to 2025. Findings reveal a technological reorientation from diversified renewables toward photovoltaic consolidation and subsequent material-driven optimisation. Assignee analysis demonstrates shifting competitive dynamics, with traditional manufacturers declining as power electronics specialists gain prominence. LLM-based thematic clustering indicates that green energy innovation is structured as a hierarchical technology stack rather than a collection of isolated fields. The MPISA framework offers a replicable methodology for understanding technological evolution and informs strategic decision-making for advancing sustainable energy transitions.
关键词
patent data analysis; green energy; sustainability; larger language model; SciBERT
KeyWord
patent data analysis; green energy; sustainability; larger language model; SciBERT
基金项目
页码 -
  • 参考文献
  • 相关文献
  • 引用本文

Cheng-Hsien Hsieh, Yu-Chao Cheng. Analyses of Green Energy Innovation Using Patent Data [J]. Journal of Sustainability Research. 2026; 8; (2). - .

  • 文献评论

相关学者

相关机构