Visualization Analysis of Knowledge Graph of Coal Mine Risk Early Warning

Abstract

To systematically reveal the development landscape and evolving trajectories of coal mine risk early warning research in China, this study combines CiteSpace-based visual analytics with bibliometric methods. A total of 688 core journal articles published in the CNKI database from 2001 to 2025 were used to construct a domain knowledge graph. The analysis examines publication trends, journal distribution, author and institutional collaboration networks, keyword co-occurrence, and burst-keyword evolution.The results show that domestic research has progressed through three phases—initial exploration, rapid expansion, and fluctuating stabilization. A stable set of core journals (e.g., Coal Mine Safety and Coal Technology) and influential authors (e.g., Jiang Fuxing and Zhang Qinghua) has gradually emerged, while overall collaboration remains limited and relatively dispersed. Four major research clusters are identified: early-warning systems, gas early warning, rock burst early warning, and coal spontaneous combustion early warning, with spontaneous combustion, indicator gases, and graded early-warning frameworks becoming prominent frontiers in recent years.Burst-keyword evolution further indicates a shift from macro-level hazard identification toward multi-source data fusion and intelligent early-warning technologies, alongside increasing integration with IoT, big data, and artificial intelligence. Overall, coal mine risk early warning research in China is moving toward more refined, intelligent, and system-oriented development, providing scientific support for smart mine construction and modernized coal mine safety governance.