地质科学大数据背景下的矿体动态建模方法探讨

三维地学建模的理论与方法在大数据时代该如何发展，是当下这一领域研究人员非常关注的问题。从现代三维地学建模的重要方法——隐式建模的角度，对三维地学建模方法与地质大数据系统的有机集成、以及如何利用大数据技术提高地学建模的效率和质量等问题进行了探讨。初步提出了一套基于地质科学大数据的三维地学建模方法和流程，包括:地质大数据的搜集、主题大数据系统的搭建、地质特征要素的深度挖掘和三维地学模型的动态构建。同时，也指出了大数据背景下高质效三维地学建模的关键在于:研究实现充分顾及地质对象和地质科学大数据特点的地学人工智能、机器学习、数据挖掘及空间推断方法。通过一个典型矿体建模的应用实例对所提方法的可行性进行了验证。 

Discussion on dynamic orebody modeling with geological science big data

As a key technology of geological information systems and an important method for geological information science, three dimensional geoscience modelling (3DGM) faces many opportunities and challenges in the age of big data. Many experts in this field are discussing ways to improve 3DGM in this context. From the point view of 3D implicit modeling, an up-to-date 3DGM method, this paper discusses several issues on this point such as the integration of a geological big data system and 3DGM, and the possible ways to enhance the quality and efficiency of 3DGM based on big data technologies. The paper proposes an initial integration framework of big data based 3DGM consisting of data collection, construction of subject-oriented big data system, geological feature mining and dynamic orebody modeling. This research also suggests that the approaches of geological artificial intelligence, machine learning, data mining, and predication should be designed and applied with full consideration of essential characteristics of the geological objects under study, which is crucial to improve the modeling result. The feasibility of the proposed method is illustrated by a case study on orebody modeling.