Multi-source data integration modeling and spatial analysis for optimal design of high-level radioactive waste geological repository
摘要
In the site evaluation, design, and construction of high-level radioactive waste (HLW) geological repositories, clear insight expression and reliable geological models are indispensable, while the spatial relationships between tunnel structures and affected zones are equally crucial for construction safety. However, traditional geological modeling methods using scarce preliminary data often suffer from loss of precision, and subsequent spatial analysis is frequently neglected. This paper presents a comprehensive methodological framework for refined modeling of the repository geological environment and tunnel structures, with particular emphasis on tunnel safety planning that incorporates spatial analysis of unfavorable geological bodies/interfaces. The geological potential prediction based on Universal Cokriging gradient field interpolation is developed using multi-source data derived from regional surveys. Safety control standards and optimization design processes are established to address the spatial layout of repository tunnels within geological bodies. The proposed framework is applied to China’s Beishan HLW repository project. Results demonstrate that data integration and modeling methods effectively enhance structural refinement. Modeling robustness is validated through systematic 10% incremental reductions in input data. Overall repository planning and layout incorporating geographic information system (GIS) principles enhance tunnel design safety and operational performance.