Stability Analysis and Optimization of Mechanized Construction Parameters for Bench Method Excavation of Large-section Soft Rock High-speed Railway Tunnels

To improve the construction quality and efficiency of mechanized bench excavation for large-section tunnels in soft rock, the Xi'an-Yan'an High-speed Railway Xin Yan'an Tunnel Project was taken as the engineering background. A three-dimensional numerical model was established using FLAC 3D to systematically compare the deformation of surrounding rock, principal stresses of primary support, distribution of plastic zones, and safety factors under four construction methods: the three-bench long-bench method, three-bench micro-bench method, two-bench long-bench method, and two-bench micro-bench method. The optimal construction method was determined by comprehensively considering construction efficiency. Furthermore, an orthogonal experimental design combined with the Analytic Hierarchy Process (AHP) was adopted to optimize the upper-bench height, upper-bench length, and excavation advance, and to investigate their effects on crown settlement, sidewall convergence, face extrusion displacement, and primary support stress, thereby determining the optimal parameter combination. The results indicate that the two-bench method performs overall better than the three-bench method, and that bench length has a significantly greater influence on surrounding rock stability and primary support stress than the number of benches. Compared with the long-bench method, the micro-bench method reduces crown settlement, invert heave, and sidewall convergence by 23.1%, 13.0%, and 24.4%, respectively, and results in smaller plastic zones and potential risk areas. In contrast, the long-bench method reduces face extrusion displacement and tensile stress in the primary support by 2.2% and 43.0%, respectively, and improves construction efficiency by 15.5%. Comprehensive analysis identifies the two-bench long-bench method as the optimal construction method. The influence of key parameters follows the order: upper-bench height, excavation advance, and upper-bench length. The optimal parameter combination is an upper-bench height of 0.6H–0.7H (where H is the tunnel height), an excavation advance of 2–3 m, and an upper-bench length of 50–60 m. Field monitoring shows that surrounding rock deformation meets code requirements, and the measured data are in good agreement with the predicted values, verifying the reliability of the model. Compared with the conventional drill-and-blast method, mechanized construction reduces surrounding rock deformation by 10%–20%, controls over-excavation and under-excavation within 15~17 cm, limits concrete consumption to 160%–170% of the design quantity, and increases construction progress by 13.3%.