“太阳石矿山大模型”针对煤矿多维度应用场景的具体需求，从基础设施层、数据资源层、算法模型层、应用服务层、安全可信与测试层、行业生态层六个层面进行建设。其中，数据资源层汇聚了煤炭行业海量的多模态数据，包括安全监测数据集500亿条、视觉图像数据集300万张、包括20本期刊与200亿字技术文献的专业学术期刊数据集等；算法模型层可支撑自主可控的100+各类AI算法；应用服务层包括矿山多模态数据交互、矿山空间重建与计算、矿山工况运维决策分析、矿山灾害预警决策分析、矿山智能设计等，能够洞察行业各类业务需求的2000+共性AI场景，能够有效降低“太阳石矿山大模型”赋能具体行业应用的难度，为下游企业提供便捷的应用方式，有助于促进太阳石大模型生态的构建。

“太阳石矿山大模型”的建设是中国煤科煤科总院全面贯彻落实党中央、国务院关于发展新一代人工智能、推进数据要素化建设的重大决策部署，助力煤炭产业数字化、智能化、绿色化转型的重要举措，是中国煤科引领煤炭智能化、数字化发展的关键一步。未来，随着大模型建设及数据治理工作的逐步推进，“太阳石矿山大模型”将提供煤炭行业所需的生成式人工智能基础应用能力，引领煤炭行业数据治理及大模型合规评测体系制定，引导行业安全、可信、绿色发展。

According to the specific needs of multi-dimensional application scenarios in coal mines, the "Sun Stone Mine large model" is constructed from six levels: infrastructure layer, data resource layer, algorithm model layer, application service layer, safety, reliability and testing layer, and industry ecological layer. Among them, the data resource layer gathers massive multi-modal data of the coal industry, including 50 billion safety monitoring data sets, 3 million visual image data sets, and professional academic journal data sets including 20 journals and 20 billion words of technical literature. The algorithm model layer can support 100+ kinds of autonomous and controllable AI algorithms; The application service layer includes mine multi-modal data interaction, mine space reconstruction and calculation, mine working condition operation and maintenance decision analysis, mine disaster warning decision analysis, mine intelligent design, etc., which can gain insight into the 2000+ common AI scenarios of various business needs in the industry, and effectively reduce the difficulty of enabling specific industry applications of "Sunstone mine large model". Providing convenient application methods for downstream enterprises is helpful to promote the construction of solar stone large model ecology.

The construction of the "Solar Stone Mine large model" is a major decision and deployment of the CPC Central Committee and The State Council on the development of a new generation of artificial intelligence and the promotion of the construction of data elements, an important measure to help the digital, intelligent and green transformation of the coal industry, and a key step for China Coal to lead the intelligent and digital development of coal. In the future, with the gradual advancement of large model construction and data governance work, "Sun Stone Mine large model" will provide the basic application capabilities of generated artificial intelligence required by the coal industry, lead the development of data governance and large model compliance eva luation system for the coal industry, and guide the safe, credible and green development of the industry.