To address the problems such as insufficient real-time data analysis, lack of intelligent decision-making support, and limited dynamic optimization capabilities of in-situ leaching dynamic mining system in practical application, the advantages of integrating Digital Twin technology with the in-situ leaching process for uranium extraction were analyzed. By incorporating advanced technologies such as muon imaging, fiber-optic water level monitoring, advanced sensing technologies, artificial intelligence, and big data analysis, an efficient digital-intelligent uranium mining platform was constructed. The platform includes geological structure model, groundwater seepage model, reactive transport model, and intelligent agent model. The core algorithms cover deep learning, data assimilation, multi-objective optimization, and uncertainty analysis. The establishment of the dynamic in-situ leaching mining system can improve the efficiency of uranium resource development and can provide valuable reference for the extraction of other mineral resources, thus having certain theoretical significance and practical value.