Amid the advancement of intelligent mining, transparent geological assurance has become a prerequisite for safe and efficient extraction of mineral resources. This study investigates multi-factor coupled high-precision mine geological modeling and its application in the prevention and control of goaf water accumulation hazards. To address accuracy limitations imposed by sparse drilling data, the Kriging spatial interpolation algorithm was employed to generate 80 virtual boreholes, increasing data density by 4.21 times and effectively overcoming the uneven spatial distribution of the original data. Based on the optimized dataset, a high-precision multi-factor coupled 3D geological model was constructed using the 3DMine platform and Triangular Irregular Network (TIN) method, integrating stratum lithology, goaf spatial morphology and groundwater dynamics. The underlying database is expanded by incorporating virtual boreholes to optimize sampling density during TIN modeling and further improve the performance of spatial interpolation. Consequently, the overall accuracy of the geological model is ultimately enhanced. As indicated by the validation results, a root mean square error (RMSE) of 0.86 and a coefficient of determination (R2) of 0.92 are achieved by the constructed model. It is demonstrated by the quantitative metrics that the actual stratigraphic structure can be reconstructed with reliable accuracy using the proposed method. Furthermore, the model can accurately characterize the 3D structural properties of underground mine spaces. By integrating the contour perpendicular method, coupling analysis of goaf morphology and groundwater migration paths is further performed and accurate calculation and visual simulation of potential groundwater flow paths within the goaf are achieved. Taking the mining panel as the basic research unit and combining actual mine production conditions, the coupling law among goaf topographic features, local hydrological boundary conditions and internal water flow field distribution was analyzed, enabling the accurate prediction of the spatial distribution of goaf water accumulation. Surface water runoff theory and hydrological modeling methods are organically integrated in this study. A reliable theoretical basis and technical support are provided for the optimal design of goaf water hazard prevention and control schemes, ensuring precise and safe coal mining and effectively addressing the challenges in groundwater hazard prevention.