智能开采工作面三维地质模型构建及误差分析

地质条件的复杂性是影响当前智能开采进一步发展的关键问题之一，亟需构建高精度回采工作面三维地质模型。通过分析智能开采地质模型的构建方法，并以黄陵一号矿某智能工作面为例，结合工作面所有的地质勘探资料，利用TIM-3D建模软件分别构建了工作面初始静态模型和回采工作面动态模型，搭载透明工作面数字孪生系统对智能开采地质模型进行展示；通过对比回采揭露真实煤厚值与地质模型预测煤厚值，分析静态地质模型与动态地质模型的误差，探讨模型误差产生的原因。分析认为：静态地质模型精度不能达到智能化开采的地质要求；更新后的动态地质模型可显著缩小煤厚预测误差，基本能达到智能化开采的地质需求；模型的误差是测量误差、采样数据量及其分布、插值算法选取共同造成的。综合认为模型的建立要充分融合工作面所有地质信息，模型建立巷道标志点的间隔应该小于10 m，模型动态更新的推采距离应该小于15 m。研究结果对于充分认识当前智能开采地质模型精度水平有重要意义，为下一步智能开采地质保障技术的发展具有借鉴意义。

3D geological model construction and error analysis of intelligent mining working face

The complexity of geological conditions is one of the key issues affecting the further development of current intelligent mining, and there is an urgent need to build a high-precision 3D geological model of the mining face. This article analyzes the construction method of the intelligent mining geological model, and takes an intelligent working face of Huangling No.1 Mine as an example, combines all the geological exploration data of the working face, and uses the TIM-3D modeling software to construct the initial static model and the working face respectively. The dynamic model of the working face is equipped with a transparent face digital twin system to display the intelligent mining geological model; by comparing the actual coal thickness value revealed by the mining and the geological model predicting the coal thickness value, the error between the static geological model and the dynamic geological model was analyzed. The reasons for the model errors were explained. The analysis concluded that: The accuracy of the static geological model cannot meet the geological requirements of intelligent mining; the updated dynamic geological model can significantly reduce the coal thickness prediction error and basically meet the geological requirements of intelligent mining; the error of the model is the measurement error and the amount of sampled data It is caused by the selection of its distribution and interpolation algorithm. It is comprehensively believe that the establishment of the model should fully integrate all the geological information of the working face, the interval between the roadway marker points in the model establishment should be less than 10 m, and the advancing and mining distance for the dynamic update of the model should be less than 15 m. The research results are of great significance for fully understanding the accuracy of the current intelligent mining geological model, and for the next step of the development of intelligent intelligent mining geological support technology.