Abstract:This study aims to systematically investigate the mechanism and effectiveness of full-section rock bolt support in stabilizing tunnel faces during excavation. Based on FLAC3D, a numerical model of a three-centered circular large-section tunnel was established. By comparing the stress distribution in surrounding rock, lining mechanical behavior, and tunnel face deformation with and without full-section rock bolt support, the working mechanism of the support system was revealed. A single-factor analysis was conducted to examine the influence of bolt spacing, length, and rock mechanical parameters on support effectiveness. The results demonstrate that full-section rock bolt support significantly enhances tunnel face stability, reducing extrusion deformation by 20.7% and effectively constraining stress release in the core soil. The bolt forces exhibit characteristic spatial distributions: axial force shows a single-peak pattern with its maximum at 0.3D ahead of the tunnel face, while shear force displays a double-peak characteristic with its zero point coinciding with the axial force peak zone, collectively identifying the potential failure surface. Parametric analysis indicates that reducing bolt spacing from 2.0 m to 0.5 m increases the support effectiveness from 6.7% to 41.7%. The improvement in support effect diminishes when bolt length exceeds 0.5D, while increased rock strength parameters reduce the relative support effectiveness. In contrast, greater tunnel depth enhances the support performance, achieving an effectiveness of 25.92% at 150 m depth. The study concludes that full-section rock bolt support constitutes a passive reinforcement system that enhances core soil stiffness through group-anchoring effects, with its performance being influenced by both bolt parameters and rock conditions. For engineering practice, it is recommended to maintain bolt spacing within 1 m and bolt length within 1D to effectively control deformation in deep-buried soft rock conditions and ensure tunnel construction safety.