Artificial intelligence-empowered coal geology: advances, challenges, and prospects
Abstract
Objective The rapid development of artificial intelligence (AI) is reshaping the research paradigm of geosciences. Coal geology focuses on coal, coal seams, coal-bearing strata, coal basins, and the mineral resources that coexist or are associated with coal-bearing strata. These research objects exhibit highly complex material compositions, multiscale characteristics, and pronounced spatiotemporal variability. Consequently, major research directions of coal geology, including coal petrology, coal quality evaluation, critical metals in coal-bearing strata, and coalbed methane exploration and development, are particularly well suited to AI-based approaches. Methods AI is driving coal geology beyond traditional, experience-based identification and localized statistical analysis toward a new paradigm integrating multi-source data fusion, intelligent prediction, and mechanistic interpretation. It also provides new opportunities for discovering previously unrecognized geological phenomena. Results This paper systematically reviews recent advances in AI applications to the intelligent identification of coal macerals, prediction of coal quality, assessment of the occurrence modes and resource potentials of critical metals in coal-bearing strata, and prediction of coalbed methane. AI technologies have significantly improved the efficiency and accuracy of maceral image recognition, coal quality parameter prediction, identification of critical metal occurrence patterns in coal, and comprehensive resource evaluation. However, several challenges remain, including the scarcity of high-quality datasets, limited cross-regional generalization capability, weak model interpretability, insufficient consistency with geological mechanisms, inadequate knowledge constraints, and limited implementation in engineering scenarios. Conclusions Future research should strengthen the development of standardized large-scale databases, promote the deep integration of geological knowledge with data-driven models, develop interpretable and transferable AI models, and establish an intelligent research paradigm oriented toward major scientific questions in coal geology and the green and efficient development of critical mineral resources. Significance This review highlights the transformative role of AI in advancing coal geology from experience-driven analysis toward knowledge-guided, intelligent research, providing methodological support for major scientific discoveries and the green and efficient development of coal-related resources.
Related
- Thermal-hydro-mechanical-chemical coupling processes across scales in CO2 storage: Practices, challenges, and prospects
- A multiple review on identification and fine evaluation of shale pores: Prospects and challenges
- Liquid organic fertilizers in soilless cultivation: a systematic review of production, performance, challenges, and future prospects
- Fine-scale characterization of dispersed-village landscapes: data challenges and methodological advances for rural landscape mapping and assessment in Pomerania, Poland
- An Artificial Intelligence Integrated Geospatial Decision-Support Framework for Scenario-Based Climate-Adaptive Land-Use Optimization and Environmental Risk Reduction
BibTeX
@article{shifeng2026artificial,
title = {{Artificial intelligence-empowered coal geology: advances, challenges, and prospects}},
author = {DAI Shifeng and XU Na and ZHU Wei and LI Pengfei and QI Anliangyu and LIU Yu and YAN Ruiwen and ZHAO Zhengfu},
journal = {Dizhi lixue xuebao},
year = {2026},
month = sep,
doi = {10.12090/j.issn.1006-6616.2026060},
url = {https://doi.org/10.12090/j.issn.1006-6616.2026060},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Artificial intelligence-empowered coal geology: advances, challenges, and prospects
AU - DAI Shifeng
AU - XU Na
AU - ZHU Wei
AU - LI Pengfei
AU - QI Anliangyu
AU - LIU Yu
AU - YAN Ruiwen
AU - ZHAO Zhengfu
PY - 2026
DA - 2026/09/01
JO - Dizhi lixue xuebao
DO - 10.12090/j.issn.1006-6616.2026060
UR - https://doi.org/10.12090/j.issn.1006-6616.2026060
AB - Objective The rapid development of artificial intelligence (AI) is reshaping the research paradigm of geosciences. Coal geology focuses on coal, coal seams, coal-bearing strata, coal basins, and the mineral resources that coexist or are associated with coal-bearing strata. These research objects exhibit highly complex material compositions, multiscale characteristics, and pronounced spatiotemporal variability. Consequently, major research directions of coal geology, including coal petrology, coal quality evaluation, critical metals in coal-bearing strata, and coalbed methane exploration and development, are particularly well suited to AI-based approaches. Methods AI is driving coal geology beyond traditional, experience-based identification and localized statistical analysis toward a new paradigm integrating multi-source data fusion, intelligent prediction, and mechanistic interpretation. It also provides new opportunities for discovering previously unrecognized geological phenomena. Results This paper systematically reviews recent advances in AI applications to the intelligent identification of coal macerals, prediction of coal quality, assessment of the occurrence modes and resource potentials of critical metals in coal-bearing strata, and prediction of coalbed methane. AI technologies have significantly improved the efficiency and accuracy of maceral image recognition, coal quality parameter prediction, identification of critical metal occurrence patterns in coal, and comprehensive resource evaluation. However, several challenges remain, including the scarcity of high-quality datasets, limited cross-regional generalization capability, weak model interpretability, insufficient consistency with geological mechanisms, inadequate knowledge constraints, and limited implementation in engineering scenarios. Conclusions Future research should strengthen the development of standardized large-scale databases, promote the deep integration of geological knowledge with data-driven models, develop interpretable and transferable AI models, and establish an intelligent research paradigm oriented toward major scientific questions in coal geology and the green and efficient development of critical mineral resources. Significance This review highlights the transformative role of AI in advancing coal geology from experience-driven analysis toward knowledge-guided, intelligent research, providing methodological support for major scientific discoveries and the green and efficient development of coal-related resources.
SN - 1006-6616
ER -