Evolution analysis and trend forecasting for low-carbon technologies in coal power based on a three-layer text mining framework

In the context of strengthening global climate governance, low-carbon technologies have emerged as a pivotal driving force for the low-carbon transition of the coal power industry. While understanding technological trends is critical for optimizing investment decisions and strategic planning in the coal power sector, systematic research on the evolution pathways of low-carbon coal power technologies (LCCPTs) remains limited. To address this gap, this study proposes a comprehensive framework for analyzing and forecasting the evolution of LCCPTs, leveraging textual data from academic papers and patents. Considering the complexity of the LCCPTs system, the framework employs a hierarchical text mining approach that integrates multiple text processing algorithms. It systematically identifies the category, topic, and content layers of the LCCPTs system based on different text granularities sequentially. Furthermore, it explores the evolutionary characteristics and development trends of the technology system across three dimensions: technology positioning, evolution pathways, and content trends, by analyzing intra-layer information and inter-layer relationships. Empirical findings reveal that LCCPTs can be categorized into five primary technological domains: fuel, combustion, carbon control, system coupling, and auxiliary, with substantial differences in evolutionary characteristics across these categories. Based on these insights, the future trend of LCCPTs is forecasted systematically. This study enriches the methods of technological evolutionary forecasting and provides valuable information references for the innovation and application of low-carbon technology in coal power industry.