Estimation of the three-dimensional in situ stress field around a large deep underground cavern group near a valley

Abstract

Abstract Understanding three-dimensional (3D) in situ stress field is of key importance for estimating the stability of large deep underground cavern groups near valleys. However, the complete 3D in situ stress fields around large deep underground cavern groups are difficult to determine based on in situ stress data from a limited number of measuring points due to the insufficient representativeness and unreliability of such measurements. In this study, an integrated approach for estimating the 3D in situ stress field around a large deep underground cavern group near a valley is developed based on incomplete in situ stress measurements and the stress-induced failures of tunnels excavated prior to the step excavation of the cavern group. This integrated approach is implemented via four interrelated and progressive basic steps, i.e. inference of the regional tectonic stress field direction, analyses of in situ stress characteristics and measurement reliability, regression-based in situ stress field analysis and reliability assessment, and modified in situ stress field analysis and reliability verification. The orientations and magnitudes of the 3D in situ stress field can be analyzed and obtained at a strategic level following these four basic steps. First, the tectonic stress field direction around the cavern group is deduced in accordance with the regional tectonic framework and verified using a regional crustal deformation velocity map. Second, the reliability of the in situ stress measurements is verified based on the locations and depths of stress-induced brittle failures in small tunnels (such as exploratory tunnels and pilot tunnels) within the excavation range of the cavern group. Third, considering the influences of the valley topography and major geological structures, the 3D in situ stress field is regressed using numerical simulation and multiple linear regression techniques based on the in situ stress measurements. Finally, the regressed in situ stress field is further modified and reverified based on the stress-induced brittle failures of small tunnels and the initial excavation of the cavern group. A case study of the Shuangjiangkou underground cavern group demonstrates that the proposed approach is reliable for estimating the 3D in situ stress fields of large deep underground cavern groups near valleys, thus contributing to the optimization of practical excavation and design of mitigating the instability of the surrounding rock masses during step excavations.