Transparent working face modelling based on TIM-3D system and its application in intelligent mining

WANG Xiaohui; ZHANG Dongliang; BAI Baojun; DU Wengang

doi:10.11799/ce202602003

Abstract：The transparent working face model serves as a core prerequisite for safe and efficient coal mine production,holding critical significance for achieving precise coal mining and enhancing geological assurance capabilities.Taking the I030903 working face of Qipanjing Coal Mine as the research object,we focus on the transparent working face model construction system and adaptive technologies for intelligent mining.Firstly,the independently developed TIM-3D(Transparent Intelligent Mine-3D)three-dimensional geological modeling system is elaborated.Relying on a true 3D modeling engine,the system integrates core technologies such as multi-source data fusion,Discrete Smooth Interpolation(DSI),and spatial mesh generation,possessing capabilities for rapid modeling of complex geological bodies,cross-platform data interaction,and seamless connection with Geographic Information Systems(GIS).Simultaneously,the TIN-GTP(Triangulated Irregular Network-Generalized Triangular Prism)algorithm and implicit iterative modeling technique are proposed,effectively resolving elevation conflict challenges in the precise modeling of reverse faults.Based on these technologies,the TIM-3D system is utilized to construct the transparent working face model,which clearly visualizes the spatial morphology and distribution patterns of strata,rock layers,coal seams,and structures within a 30-meter range above and below the No.9 coal seam.The model enables bidirectional exchange of information with the shearer,armored face conveyor,and hydraulic supports.Through a deep reinforcement learning algorithm(DQN-NAF),cutting trajectories are generated and dynamically optimized,forming a closed-loop control process for intelligent mining,thereby significantly improving mining efficiency and safety.Future research will deepen focus on real-time IoT monitoring,optimization of machine learning algorithms,and the construction of an intelligent decision-support platform to further advance the intelligence level coal industry.