In a high geo-temperature environment, it is rarely reported that geo-temperature has been considered during a back analysis. This may cause the initial geo-stress field that is obtained by a back analysis to be wrong. In this study, according to the theory of elasticity, the theoretical solution of the hydraulic fracturing equation is obtained in a high geo-temperature environment. Since the vertical stress that is obtained by the hydraulic fracturing method is calculated using the density of overlying strata, this vertical stress lacks the thermal stress that is caused by geothermal gradients. Therefore, in a high geo-temperature environment, inverting the initial geo-stress field of rock masses directly using the stress that is measured by the hydraulic fracturing method can cause serious errors. We propose that the regression coefficient of a gravitational stress field should be set to one during a back analysis if stresses are measured by the hydraulic fracturing method, and this regression coefficient should not be equal to one if stresses are measured by overcoring methods. We also propose a workflow for the back analysis of the initial geo-stress field of rock masses that considers geo-temperature, and this workflow is applied to the Sangzhuling tunnel in China.

Due to the limitations of the terrain and other reasons, twin-tunnels with small clear spacing are becoming more and more common. The reasonable clear spacing of twin-tunnels in weak surrounding rock is an urgent problem. In this paper, based on pressure-arch theory (PAT), the method used to calculate the loosening pressure on the surrounding rock of twin-tunnels under three cases was derived by considering the additional disturbance of the excavation on the middle rock pillar through the amplification factors of the sliding angles, k₁ and k₂. Taking the weak surrounding rock as an example, we discussed the effect of the clear spacing variation on the loosening pressure of the surrounding rock, and obtained the dangerous clear spacing of twin-tunnels. The accuracy and applicability of this approach were verified by the in-site data that were measured. Considering the limitation of the pressure-arch theory on a deep-buried tunnel under high geo-stress, based on the Jiuzhai valley-Mianyang highway in China, the reasonable clear spacing of twin-tunnels in weak surrounding rock at different buried depths, i.e., 400 m–1,600 m, is discussed using the numerical simulation method. The analytical solution indicated that the loosening pressure of the inner side of the first tunnel was greater than the one of the second due to the amplification factors of the sliding angles, k₁ and k₂. The in-site data showed that in the weak surrounding rock mass, when k₁ and k₂ are taken as 1.15 and 1.3, respectively, the loosening pressure distribution law is closest to the actual situation.

The inversion of the ground stress field in rock masses is critical for accurate tunnel and underground engineering design. This study addresses the challenge of accurately capturing both the primary and secondary stress field components in rock masses. The ground stress field consists of the primary stress field, generated by applied tectonic loads, and a secondary stress field, which cannot be fully explained by these loads and is attributed to long-term tectonic processes. This unexplained secondary stress field is often non-random in nature. To improve the accuracy of the ground stress field inversion, we propose prioritizing the use of a regression model with a constant term. This model better accounts for the secondary stress field by capturing long-term tectonic influences. The constant term in the regression model is shown to represent the non-random secondary stress field, which cannot be explained by applied tectonic loads. Furthermore, we define two key conditions for applying this regression model: (1) the constant term should not exceed the maximum measured stress and preferably should not surpass the minimum measured stress, and (2) the residual sum of squares of the regression model with a constant term should be smaller than that of the model without a constant term. By incorporating the constant term, the model improves the representation of both primary and secondary stress fields, offering a more accurate inversion of the ground stress field, especially when the stress field contribution from independent variables is incomplete.

In foundation pit engineering, precise determination of the physical–mechanical parameters of the surrounding rock is essential for reliable simulation of rock deformation and anchor cable forces. A foundation pit engineering project in Shapingba District, Chongqing, was selected as a case study. A numerical model was developed using FLAC3D, and 64 working conditions were designed via orthogonal experiments to serve as training samples. Global optimization inversion of the samples was performed using a BP neural network enhanced by particle swarm optimization. Using selected monitoring data of surrounding rock displacement and anchor cable forces, inversion was conducted to determine the physical–mechanical parameters of the foundation pit surrounding rock, and the FLAC3D model inputs were subsequently updated. Finally, simulated results were validated against field measurements. The maximum relative error of surrounding rock displacement reached 8%, with only 3% at the pit center. The largest settlement occurred in the eastern section, where the relative error was 5%. For anchor cable forces, the maximum relative error was 7.9%. This study employed a PSO-BP neural network to invert the physical–mechanical parameters of the foundation pit surrounding rock and introduced a two-stage validation using measured displacements and anchor cable forces. The approach enhances inversion accuracy and provides a practical reference for similar foundation pit engineering applications.