Effectiveness of Advanced Support at Tunnel Face in ADECO-RS Construction

Abstract

Tunnel construction in weak and fractured strata often faces risks such as tunnel face instability and large deformation of surrounding rock, which are difficult to effectively control using conventional support methods. Based on the engineering background of the No. 8# TA Tunnel in the F3 section of Georgia’s E60 Highway, this study employed ADECO-RS and developed a 3D numerical model with finite difference software to simulate full-face tunnel excavation process. The influence of advanced reinforcement measures on the stability of the surrounding rock was systematically investigated. The control effectiveness of different advanced reinforcement schemes was evaluated by comparing the displacement field, stress field, and plastic zone distribution of the surrounding rock under three conditions: no support, advanced pipe roof support, and a combination of pipe roof and glass fiber bolts. A comprehensive quantitative analysis of the synergistic effect of the combined reinforcement was also performed. The results indicated that significant extrusion deformation of the tunnel face and vault settlement occurred after excavation. The pressure arch developed within a range of 17.5 to 22 m above the tunnel vault. The surrounding rock of this tunnel was classified as type B (short-term stable). Deformation primarily occurred within one tunnel diameter ahead of the face, with the deformation rate significantly reduced after support. Advanced pipe roof support effectively restrained surrounding rock deformation, while the combination of advanced pipe roof and glass fiber bolts delivered better performance: reducing final convergence by 73.10%, pre-convergence by 82.69%, and face extrusion by 87.66%. The combined support also contracted the pressure arch boundaries from 17.5 to 22 m to 6–12.5 m, reduced the extent of major principal stress deflection, and significantly shrinks the plastic zone. Glass fiber bolts played a key role in controlling plastic zone expansion and ensuring stability. This study provides theoretical and numerical references for safe construction and advanced support design in tunnels under complex geological conditions.

1. Introduction

With the acceleration of global urbanization and the increasing density of transportation networks, tunnel engineering, as critical infrastructure for overcoming geographical barriers and intensively utilizing underground space, is experiencing unprecedented growth in both scale and quantity. However, tunnel construction often encounters complex and variable geological conditions. Instabilities at the tunnel face, large deformations of the surrounding rock, and collapse hazards frequently occur due to adverse geology such as weak, fractured, and water-rich strata. For example, during the construction of the Tartaiguille Tunnel on the Lyon-Marseille high-speed railway in France, which traverses a fractured rock mass, the surrounding rock underwent severe deformation. The vertical convergence reached 150 mm, causing severe cracking and spalling of the shotcrete. This necessitated frequent removal of the support structure, posing serious threats to personnel safety and resulting in significant project delays and substantial financial losses [1]. As another example, on 16 November 2021, a severe collapse accident occurred in the Yantangshan Tunnel on the Hangzhou-Quzhou Railway. The collapsed rock mass weighed approximately 10 tons, resulting in three fatalities, direct economic losses of 4.374 million yuan, and a severe negative social impact. The fundamental cause of the aforementioned geohazards lies in the intense spatial variability and extreme uncertainty of the geological conditions [2]. This spatial variability leads to a highly heterogeneous distribution of the surrounding rock mass’s mechanical parameters, resulting in poor self-stabilization capacity and significant challenges in its prediction. Conventional support methods, however, primarily focus on controlling deformation within the excavated tunnel sections. They fail to fundamentally and proactively improve the mechanical properties and stability of the unknown and highly variable ground ahead of the tunnel face. Therefore, an in-depth investigation into the mechanism and effectiveness of face reinforcement measures used in the ADECO-RS is imperative. This research, centered on the principle of active deformation control, represents not only a direct response to demanding ground conditions but also a critical endeavor to address the core challenges arising from geological spatial variability. It holds substantial theoretical value and practical significance for overcoming the difficulties of tunnel construction in poor ground conditions and for enhancing disaster prevention and control capabilities.

The analysis of the mechanism and effectiveness of advanced tunnel face reinforcement has consistently been a research hotspot in the field of tunnel and underground engineering. Numerous scholars worldwide have conducted extensive and productive research on this topic [3,4,5,6,7]. In terms of theoretical analysis, Yi et al. [8], based on an established rockbolt–soil interaction model, used the extrusion deformation of the advanced core soil as a parameter to analyze the reinforcement mechanism under conditions of coupled deformation between rockbolts and soil. They also proposed using a reinforcement coefficient ‘k’ to describe the effectiveness of reinforcement under such coupled deformation conditions. Zhang et al. [9] developed a load-structure model for the pipe roof based on Pasternak’s elastic foundation beam theory. They analyzed the mechanical behavior of the advanced pipe roof during tunnel excavation and noted that increasing the strength of the surrounding rock can significantly enhance the support effectiveness of the pipe roof. Rong et al. [10] established an arc-axis theoretical model for the soil arching effect between steel pipes and, by integrating Pasternak’s elastic foundation beam theory, analyzed the deformation mechanism of a tunnel advanced pipe roof under the pressure of the surrounding rock. Regarding field monitoring, Zhang et al. [1] applied the ADECO-RS approach to the design and construction of the support system for the Tartaiguille Tunnel. Through field monitoring, they subsequently analyzed the post-construction tunnel convergence deformation and the extrusion deformation of the advanced core soil at the tunnel face. Sun et al. [11], using the Baojiang Road–Heilongjiang Road section tunnel of Qingdao Metro Line 3 (Phase I) as a case study, adopted field tracking monitoring to analyze the influence of ADECO-RS construction on tunnel deformation and surface settlement. Wang et al. [12], based on the engineering case of the 106 Mining Panel in the Yuandian No. 2 Mine, successfully controlled roadway roof collapse by employing a combined support technology of advanced grouting with grouted rock bolts and grouted anchor cables. Monitoring results indicated that this technique effectively restrained the deformation of the surrounding rock, demonstrating significant support effectiveness. Zheng et al. [13] conducted field tests, including in situ stress measurement, loosened zone measurement, and deformation monitoring. Through these, they investigated the large deformation failure mechanism of the Wanhe Tunnel on the China-Laos Railway and evaluated the support effectiveness of the advanced pipe roof in controlling the tunnel’s deformation. The aforementioned research provides important references for the study of advanced tunnel face reinforcement. However, theoretical analyses often highly idealize complex geological and construction processes. Their theoretical models struggle to accurately describe the heterogeneity of the rock mass and the interactions involved in reinforcement. Furthermore, theoretical analysis cannot precisely simulate the spatiotemporal effects inherent in dynamic construction, often leading to discrepancies between predicted outcomes and actual engineering effectiveness. While field monitoring can accurately reflect the mechanical behavior of the surrounding rock during advanced reinforcement, it is limited by the intricacy of in situ geological conditions and the availability of monitoring points. Consequently, in situ testing finds it difficult to comprehensively obtain full-field, multi-dimensional information on the surrounding rock throughout the excavation and advanced reinforcement process. With the maturation of numerical simulation methods, their advantages in terms of cost-effectiveness, precision, and controllability are leading increasingly more scholars to adopt them for investigating issues related to advanced tunnel face reinforcement. Yang Yuan et al. [14] employed a combined approach of numerical simulation and field monitoring to research the advanced pipe roof effectiveness in controlling surrounding rock deformation for a high-speed railway tunnel. Their study investigated various scenarios involving different thicknesses of soft plastic loess layers and their relative positions to the tunnel. Jiang et al. [15], using the Tianshan Tunnel as a case study, utilized finite difference software to perform numerical simulations of different excavation schemes for the shallow-buried tunnel. The analysis examined the support mechanism of advanced pipe roof and the influence of different construction plans on its support effectiveness. Yuan et al. [16], based on the shield tunneling departure project of Guangzhou Metro Line 13, applied an improved BQ method to classify the tunnel surrounding rock. They then utilized finite element software to assess the influence of different parameters of the pipe roof-such as its diameter, number of pipes, grouting layer thickness, and layout range-on its support effectiveness. Liu et al. [17] established a numerical model to systematically investigate the evolution patterns of surface settlement, vault settlement, and internal forces of the support structure during tunnel excavation and support under various conditions. Their study revealed the mechanism of advanced small pipe roof support and further developed an analytical model for evaluating the effectiveness of pre-support. To investigate the performance of six advanced support methods, Li et al. [18] established a FLAC3D 6.0 model to simulate tunnel excavation in a sandy cobble stratum. They obtained the displacement and stress field distributions of the soil around the cavity for each scenario and conducted a systematic analysis of the control effectiveness and influencing mechanisms of different advanced reinforcement methods on tunnel stability. Wang et al. [19] employed the finite difference method to conduct a comparative analysis of the multi-field response characteristics of the surrounding rock under different advanced support measures. They systematically evaluated the effectiveness of these measures in enhancing stability during the tunnel excavation process. The aforementioned research findings have significantly advanced the study on the effectiveness of advanced tunnel face reinforcement and provided crucial theoretical support for the safe construction of underground engineering under adverse geological conditions.

Current research on advanced reinforcement primarily focuses on analyzing the effectiveness of individual support measures, while studies on the advanced pipe roof, glass fiber bolts, and particularly the combined effectiveness of these two advanced reinforcement methods are relatively scarce. Furthermore, while existing studies have predominantly focused on the deformation response of the surrounding rock and tunnel structures, research on the pre-deformation behavior of the surrounding rock induced by pre-reinforcement measures remains relatively limited. Therefore, grounded in the theoretical framework of ADECO-RS and using the 8# TA Tunnel in Section F3 of the Georgia E60 Highway as the case study, this research employs finite difference software to establish a three-dimensional numerical model to simulate the tunnel excavation process. By analyzing the mechanical response of the surrounding rock during the excavation process, its stability type was evaluated. The mechanical response characteristics under different advanced reinforcement measures were further compared to systematically assess the effectiveness of various advanced support schemes. A comprehensive quantitative analysis of the synergistic effects of combined advanced reinforcement was conducted. This study reveals key mechanical mechanisms, thereby transcending a simple engineering case summary. Ultimately, it provides generalizable quantitative conclusions and a deeper mechanistic understanding of the role of combined advanced reinforcement, offering a valuable scientific reference for the optimal design of tunnel support schemes under similar unfavorable geological conditions.

2. Project Profile

The E60 Highway in Georgia traverses the entire country, extending from the border with Azerbaijan in the east to the Black Sea port of Poti in the west. It primarily crosses three geomorphological units: the Zemo Imereti Plateau, the Colchis Piedmont Rolling Zone, and the Colchis Alluvial Plain, making it a vital section of the Asia-Europe transportation corridor. The Section F3 project of the Georgian E60 Highway is located in the Imereti region, specifically the Shukruti to Zestafoni section of the E60 expressway. The total length of the route is 6.768 km, which includes five tunnels totaling 6.642 km in length, (see Figure 1a). The route primarily traverses the Zemo Imereti Plateau, which is the highest-altitude area in central Georgia. This area exhibits extremely complex geological conditions, featuring multiple developed faults and fracture zones, with widespread adverse geological structures along the route.

The location of the 8# TA Tunnel in the Georgia E60–F3 section is shown in Figure 1a. The tunnel has a total length of 1171 m with a maximum burial depth of approximately 170 m. During construction, the tunnel traverses three fracture zones and primarily passes through the J₂b₂B sub-unit formation (as shown in Figure 1b). The lithology of this formation is predominantly quartz-porphyry, with developed porphyry, porphyry breccia, lava breccia, layered tuff, tuff, and tuff breccia. The rock type is primarily hard rock. The weathering products are leached and colluvial deposits, with grain sizes ranging from clay to gravel and cobbles. The geological profile is shown in Figure 1b. According to advanced geological forecasts, a deeply buried section of the 8# TA Tunnel exhibits a type B2V cross-section over a length of 43.5 m. In this section, the surrounding rock mass is fractured, with disordered structural plane development and nearly zero self-stabilization capacity. Excavation in this zone is highly prone to engineering disasters such as tunnel face collapse, roof collapse, and water inrushes with mud outbursts (Figure 1c–e).

3. Mechanical Response Characteristics and Stability Assessment of Surrounding Rock

3.1. Numerical Model

The three-dimensional numerical model was developed based on the construction cross-section dimensions and geological characteristics of the 8# TA Tunnel of the Georgia E60 Highway. The center of the tunnel was selected as the coordinate origin, with the tunnel excavation direction defined as the Y-axis. The direction perpendicular to the Y-axis within the horizontal plane was set as the X-axis, while the vertically upward direction was defined as the positive Z-axis. The clear distance between the vault and invert of the model is 11.9 m, with a lateral span of 12.9 m. Based on the Saint-Venant principle, a model dimension of five times the tunnel diameter was adopted to minimize boundary effects [20,21]. The three-dimensional numerical model extends 120 m in the X-direction, 60 m in the Y-direction, and 90 m in the Z-direction (see Figure 2a). The model is modeled by hexahedral mesh solid elements, which are divided into 59,760 elements and 60,780 nodes. At the same time, the mesh is refined around the tunnel excavation area to accurately capture stress concentration and deformation gradient. The element size of this critical region is set to be much smaller than the expected plastic zone and pressure arch range, thus ensuring the calculation accuracy. The mechanical convergence criterion is adopted in the numerical calculation, which is the standard and strict tolerance to ensure the balance of each step in the finite difference analysis of geotechnical engineering. The general Mohr–Coulomb constitutive model is used in the numerical model of surrounding rock, and its parameters can be reliably obtained by conventional test, which can provide a stable reference for the comparative analysis of this support scheme, assuming isotropic homogeneous material properties while neglecting geological structures and groundwater effects. The physical and mechanical parameters of the surrounding rock are detailed in

Table 1. Concerned parameters of surrounding rock and primary support.

	Unit Weight (kN/m3)	Elastic Modulus (GPa)	Cohesion (MPa)	Poisson’s Ratio	Internal Friction Angle (°)
Surrounding rock	25.1	1	0.33	0.3	22
Primary support	25.0	29	-	0.25	35
Grout	25.0	20	1	0.25	35

. In order to prevent the rigid body displacement and simulate the in situ stress field, the boundary constraint of the model is set as the bottom is completely fixed, the left and right sides are constrained by the X-direction displacement, the front and back ends are constrained by the Y-direction displacement, and the top is set as the free surface. According to the burial depth, the weight of the overlying rock mass was converted into a vertical load applied to the upper boundary of the model. The calculated pressure applied to the top boundary was 5 MPa (see Figure 2b). The simulation of tunnel excavation clearly considers the sequential construction process. The whole section method is used to excavate and advance in stages. Each construction cycle simulates a footage of 1 m.

3.2. Mechanical Response Characteristics of Surrounding Rock

3.2.1. Displacement Field Response Characteristics

After the tunnel was excavated to 30 m, we created nephograms of the vertical displacement and axial extrusion displacement of the surrounding rock at this cross-section, as shown in Figure 3a,b.

Figure 3c,d illustrate the pre-convergence and pre-extrusion deformation, respectively, in the surrounding rock ahead of the tunnel face. As can be seen from Figure 3a,b, tunnel excavation induced vault settlement and floor heave deformation of the surrounding rock toward the tunnel interior. Simultaneously, extrusion displacement occurred along the axis behind the tunnel face. The closer to the excavated boundary, the more significant the vertical deformation of the surrounding rock; the nearer to the center of the excavation face, the greater the extrusion deformation of the surrounding rock. As shown in Figure 3c,d, the deformation of the surrounding rock behind the tunnel face at the 30 m excavation section can be divided into three zones: rapid increase, gradual increase, and stabilization. Within approximately one tunnel diameter ahead of the excavation face, both the vertical convergence displacement and extrusion displacement of the surrounding rock increase dramatically. The influence of excavation disturbance on surrounding rock displacement weakens with increasing distance from the tunnel face, accompanied by a gradual reduction in the rate of displacement increase. At approximately two tunnel diameters from the face, the increments in vertical convergence and extrusion displacement of the surrounding rock diminish significantly. Particularly for the extrusion deformation, the disturbance-induced deformation becomes negligible. Both pre-convergence and pre-extrusion displacements ahead of the tunnel face reach decimeter-level magnitudes. The substantial deformations occurring near the tunnel face may indicate potential spalling of partial surrounding rock layers.

Figure 4 shows nephograms of vertical convergence and extrusion displacement of the surrounding rock ahead of the tunnel face when the tunnel was excavated to 10 m, 20 m, and 30 m, respectively. As shown in the figure, due to excavation disturbance, the surrounding rock undergoes inward convergence deformation and extrusion deformation toward the rear of the tunnel axis. The extrusion deformation of the surrounding rock becomes more pronounced closer to the center of the tunnel face. As the excavation depth increases, the extrusion displacement of the tunnel face progressively intensifies. The vertical displacement at the vault of the tunnel face generally exceeds that at the invert, with both magnitudes increasing alongside advancing excavation depth. Compared to the extrusion deformation, the vertical deformation is confined to a specific range ahead of the tunnel face, forming a collapse arching effect. Both the extrusion deformation and vault settlement of the surrounding rock at the tunnel face reach decimeter-level magnitudes.

3.2.2. Stress Field Response Characteristics

The arching effect refers to the phenomenon where stress redistribution occurs in the surrounding rock due to wall unloading after tunnel excavation, forming an annular high-stress zone with enhanced bearing capacity around the tunnel. This effect primarily develops in the vault area, where an effective pressure arch can transfer the overlying load to the stable rock masses on both sides, significantly influencing the stability of the surrounding rock. This study focuses on the stress distribution above the tunnel vault. By analyzing the inner and outer boundaries of the pressure arch, the effectiveness of its self-bearing capacity is preliminarily determined. The study defines the location of the maximum value of the maximum principal stress (σ_max) in the surrounding rock as the inner boundary of the pressure arch, while the point where the maximum principal stress deflects is identified as the outer boundary [22].

Due to FLAC3D’s convention defining tensile stresses as positive, which contradicts geomechanics sign conventions, this study specifically defines the σ_max as the compressive stress with the largest absolute value. The principal stress vector diagram of the surrounding rock before excavation is shown in Figure 5a, where green represents the minimum principal stress, red indicates the intermediate principal stress, and pink segments denote the σ_max. The length of each segment corresponds to the magnitude of the absolute stress value. Figure 5b shows the principal stress vector diagram of the surrounding rock at the cross-section when the tunnel is excavated to 30 m. As shown in the figure, in the area adjacent to the tunnel vault, the σ_max is oriented horizontally, indicating significant influence from excavation disturbance in this region. As the distance between the surrounding rock at the vault and the excavation boundary increases, the direction of the σ_max gradually returns to the initial in situ stress orientation. This demonstrates that at a certain distance from the tunnel, the surrounding rock begins to form a pronounced radial load-bearing arching effect. Concurrently with the formation of this load-bearing arch, redistribution of principal stresses occurs along the tunnel axis within the surrounding rock. This is specifically manifested by significant rotation of the σ_max direction in the region proximate to the excavation face.

When the tunnel was excavated to 30 m, the nephogram of the σ_max at this cross-section is shown in Figure 6a, while the path curves of horizontal and vertical stresses in the surrounding rock at different distances from the vault are presented in Figure 6b. As can be seen from the figures, the σ_max in the surrounding rock at the vault and invert decreases, with the magnitude of stress reduction at the vault increasing as the distance to the excavation free surface decreases. In contrast, the σ_max in the surrounding rock at the haunches increases, showing relatively significant stress concentration (see Figure 6a). At the tunnel vault, the horizontal stress in the surrounding rock shows a decreasing trend, but with a smaller attenuation magnitude compared to the vertical stress. This occurs because after tunnel excavation, the surrounding rock at the vault undergoes unloading rebound deformation due to the release of radial constraints, resulting in the liberation of accumulated elastic energy in the rock and consequently forming a stress reduction zone. Meanwhile, as described in Figure 6b, the direction of the σ_max deflects at the intersection point of the horizontal and vertical stress curves. This phenomenon marks the transition where the σ_max changes from vertical to horizontal direction, indicating the outer boundary of the pressure arch. The figure shows that the inner and outer boundaries of the pressure arch are located 17.5 m and 22 m away from the center point of the tunnel face, respectively.

Figure 7 shows the stress distribution in the surrounding rock ahead of the tunnel face at different excavation depths. As observed from the figure, at excavation depths of 10 m, 20 m, and 30 m, the stress in the surrounding rock at the tunnel face, above the tunnel vault, and below the tunnel invert decreases, indicating a stress release phenomenon. This suggests a high potential for tunnel instability failure, necessitating additional control measures during construction to manage the surrounding rock.

3.2.3. Plastic Zone Response Characteristics

To investigate the distribution characteristics of the plastic zone in the surrounding rock at different tunnel excavation depths, the nephograms of plastic zone distribution at and ahead of the face under various excavation depths were extracted (see Figure 8). The variation in the plastic zone in the surrounding rock shown in Figure 8 indicates that tunnel excavation causes the surrounding rock to lose its original displacement constraints and supporting forces, disturbing the original balance of the rock mass. This results in convergent displacement of the surrounding rock toward the tunnel interior and induces plastic deformation in the rock ahead of the tunnel face. The distribution characteristics of the plastic zone exhibit approximate symmetry along the tunnel’s central axis. Moreover, a substantial plastic zone is also present in the unexcavated surrounding rock within a certain range ahead of the tunnel face. Meanwhile, as described in the figure, the surrounding rock at the tunnel face primarily undergoes tension–shear failure, while shear failure dominates at the vault and haunches, and tension–shear failure occurs at the invert. As the excavation depth increases, the extent of tension–shear failure in the surrounding rock plastic zone moderately expands.

3.3. Stability Assessment of Surrounding Rock

According to the ADECO-RS theory, the deformation induced by tunnel excavation is primarily attributed to three components: surrounding rock pre-convergence deformation, surrounding rock convergence deformation, and tunnel face extrusion deformation (see Figure 9a). The core of this theory lies in achieving overall stability of the surrounding rock by controlling the mechanical state of the advanced core soil ahead of the tunnel face. Tunnel excavation causes the surrounding rock to transition from a triaxial stress state to a biaxial stress state, wherein the deformation characteristics of the advanced core soil play a decisive role, directly influencing the surrounding rock stability. Based on the stress state after excavation, the ADECO-RS theory classifies tunnel face stability into three categories: type A (stable), type B (short-term stable), and type C (unstable). The theory emphasizes that the construction phase represents the critical period when the structure is subjected to maximum stress. During this stage, the initial stress field of the surrounding rock undergoes redistribution, causing stress flow lines to deviate near the tunnel excavation boundary and form a stress concentration zone, known as the “arching effect” (see Figure 9b). The formation and spatial positioning of this arching effect are crucial for effectively controlling the deformation response of the surrounding rock. Corresponding to the ADECO-RS tunnel face stability classification, the arching effect is also categorized into types A, B, and C, which respectively represent different states of surrounding rock stability.

In type A (stable) surrounding rock, the rock in front of the face after excavation remains primarily in an elastic state and possesses sufficient bearing capacity. Consequently, the unexcavated tunnel face (or the surrounding rock ahead) exhibits extremely minimal deformation and demonstrates high self-stability. Such rock masses may only undergo centimeter-scale elastic deformation or experience sporadic and isolated rockfalls. Owing to the high self-stability and negligible the tunnel face deformation, the arching effect can form and function effectively in close proximity to the excavation boundary (see Figure 9c). In type B (short-term stable) surrounding rock, the rock ahead of the face transitions into an elastoplastic state after excavation, exhibiting limited bearing capacity. Consequently, the tunnel face undergoes elastoplastic deformation toward the tunnel cavity while still maintaining basic stability in a short time. The surrounding rock may experience decimeter-scale deformation or localized spalling. If support measures are not implemented promptly, continuous stress release in both radial and longitudinal directions will occur, leading to progressive expansion of the plastic zone ahead of the tunnel face. The development of this plastic zone undermines the bearing capacity of the near-field surrounding rock, forcing the formation position of the “arching effect” to shift deeper into the rock stratum farther from the excavation boundary (see Figure 9d). In type C (unstable) surrounding rock, the rock ahead of the face rapidly transitions into a failure state or becomes unstable along potential sliding surfaces, with its self-bearing capacity fundamentally lost. Consequently, the tunnel face exhibits severe and uncontrolled deformation, becoming highly unstable. Due to the extensive failure or sliding of the surrounding rock, it can no longer form an effective natural arching effect (see Figure 9e).

In Section 2, through the excavation simulation of the actual tunnel conditions, it was determined that the tunnel face extrusion deformation reached 382.82 mm and the tunnel vault settlement deformation was 112.44 mm, both indicating decimeter-scale deformation of the surrounding rock. The formation state of the pressure arch after excavation was simulated and analyzed by examining the transformation patterns of the maximum and minimum principal stresses as well as the horizontal and vertical stresses in the vault region. The analysis concluded that the pressure arch of this project formed at a certain distance from the tunnel face boundary, with the inner and outer boundaries located 17.5 m and 22 m away from the center of the tunnel face, respectively. Using the ADECO-RS classification system, the tunnel face surrounding rock was categorized as type B based on its stability characteristics. As evidenced by the aforementioned analysis of different rock types, effective advanced reinforcement measures must be implemented prior to excavation for both type B and type C surrounding rock to prevent instability of the tunnel face and tunnel structure, thereby maintaining their stable conditions.

4. Analysis of Advanced Reinforcement Effectiveness

Through computational modeling and analysis in Section 3, the stability classification of the surrounding rock was determined as type B. Based on commonly used stabilization measures in the ADECO-RS and typical pre-support techniques, combined with the support applicability conditions for the geological characteristics of the tunnel’s B2V section, it was decided to employ three support schemes listed in

Table 2. Support scheme.

Conditions	Support Type
1	Primary support
2	Primary support + Advanced pipe roof
3	Primary Support + Advanced pipe roof+ Glass fiber bolts

for comparative simulation analysis. Simulations were conducted separately to analyze the pre-convergence displacement, convergence displacement, and tunnel face extrusion displacement during the excavation process, along with investigating the distribution of the plastic zone and variations in the stress field.

4.1. Tunnel Face Advanced Reinforcement and Its Function in the ADECO-RS Approach

The essence of advance pipe roof support lies in the artificial enhancement of the “arching effect”. Prior to excavation, closely spaced steel pipes are inserted into the vault surrounding rock to form a steel arch frame. Grouting is then performed to consolidate the surrounding fractured rock and soil into an integral unit, collectively forming a reinforced composite arch structure. Through the high load-bearing capacity of its shed-frame support system, the advance pipe roof effectively mitigates the adverse effects of overlying rock pressure on the tunnel and the tunnel face. The advance-installed pipe roof support system significantly enhances tunnel stability via multi-scale synergistic effects. In the longitudinal dimension, by utilizing the steel frame near the tunnel face and the stable rock mass ahead as support points, it forms an integrated beam-like structure. This effectively restrains the surrounding rock loosening load and deformation from the tunnel face extrusion zone to the ahead pre-convergence zone, demonstrating a significant spatial restraint effect (see Figure 10a). As shown in Figure 10b, in the transverse dimension, the closely arranged pipe roofs form small arch units that couple and superimpose with each other, synergistically combining with the steel arch frame to construct an integrated arch-shell layer with high load-bearing capacity. This layer serves as the primary load-bearing structure to efficiently transfer and redistribute the overlying rock pressure. Simultaneously, the cement slurry injected through the pipe walls penetrates and diffuses into the surrounding rock fractures, significantly enhancing the strength and stiffness of the Surrounding rock. Thereby, it effectively suppresses deformation and collapse risks induced by excavation disturbance and stress redistribution, while reducing the extent of the rock loosening zone. This system comprehensively embodies a coupled reinforcement mechanism integrating beam-like load-bearing, arch-shell force transmission, and surrounding rock strengthening.

The Glass fiber bolts exhibit high tensile strength, corrosion resistance, and cuttability. In tunnel construction, they do not require removal and can be directly cut and extended with new bolts for reinforcement, thereby significantly enhancing construction efficiency in large-section tunnels. The glass fiber bolts are implanted into the advance core soil through drilling, where they restrict the extrusion deformation of the rock via anchoring action (as shown in Figure 11a). From the perspective of holistic reinforcement mechanism, the bolts enhance the cohesion, internal friction angle, and uniaxial compressive strength of the advance core soil through grouting, thereby shifting the Mohr–Coulomb strength envelope upward. Simultaneously, the restraint provided by the bolts on extrusion deformation effectively inhibits excessive attenuation of the minimum principal stress, promoting the transition of the rock mass stress state from the potential failure zone (Curve 2) to the stable zone (Curve 3). Furthermore, the anchor bolts transfer the stress from the tunnel face to the deeper surrounding rock, enhancing the load-bearing capacity of the advance core soil and facilitating the formation of the arching effect. This effectively suppresses both the pre-convergence and convergence deformations of the surrounding rock (see Figure 11b). The interaction mechanism between anchor bolts and soil demonstrates that glass fiber bolts primarily suppress stress release and deformation of the surrounding rock through interface shear stress, thereby enhancing its stability. Specifically, in the anchored segment near the tunnel face, the direction of shear stress points toward the tunnel face to restrain deformation of the surrounding rock, while in the pull-out segment farther from the tunnel face, the shear stress direction points toward the deeper surrounding rock. The zero-shear stress point (neutral point) corresponds to the location where the relative displacement between the bolt and the surrounding rock is zero. At this point, the axial force in the bolt reaches its maximum and decays to zero toward both ends (see Figure 11c). To analyze its mechanical mechanism, an infinitesimal element dx within the anchored segment is selected to establish a mechanical model. It is assumed that prior to bolt reinforcement, the soil exhibits a certain amount of in situ deformation du. After reinforcement, owing to the interplay between the bolt and the soil, the internal stress within the soil increases, and the corresponding deformation decreases. Overall, the bolt effectively restrains the deformation of the advance core soil through interface shear stress.

The installation of advance support follows a meticulously designed sequence to maximize the synergistic interaction of its components. First, advance pipe roofs are installed from within the already-supported section. Following their placement, primary grouting is performed, allowing the grout to permeate the surrounding fractures and form a consolidated, load bearing “arch shell” in the crown area. This arch shell serves to pre-stabilize the excavation profile and provide a protected environment for subsequent face operations. Once the pipe roof system is in place, the tunnel face is advanced. Glass fiber bolts are then installed radially into the core soil. Grouting of these bolts further enhances the strength and integrity of the core, effectively creating a reinforced soil block. This reinforced core directly resists extrusion deformation of the tunnel face while simultaneously providing essential longitudinal and lateral support to the ends of the pipe roofs, thereby anchoring the entire protective arch.

4.2. Design of Tunnel Face Advance Reinforcement Scheme

4.2.1. Parameter Design and Modeling of Advanced Pipe Roof

According to the field construction layout, the advance pipe roof support adopted in Tunnel 8# utilizes hot-rolled seamless steel tubes with a diameter of 114.3 mm and a wall thickness of 10 mm. Each steel tube has a length of 15 m with a 5 m overlap, segmented into 6 m and 9 m sections connected by threaded couplings. Grouting holes with diameters of 10–16 mm are drilled on the pipe roof tubes at a spacing of 15 cm in a staggered pattern. A non-perforated grout-sealing segment of 150 cm is reserved at the tail end of each tube (see Figure 12a). The pipe roofs are installed on each connected steel arch frame with a circumferential spacing of 38 cm, arranged over a 140° range of the vault, totaling 39 tubes. A schematic layout of the pipe roof arrangement is shown in Figure 12b.

This section employs beam structural elements in FLAC3D to establish the numerical model of the pipe roof. The geometric dimensions and layout configuration of the pipe roof are defined according to the field construction conditions. The numerical model is shown in Figure 13, and the mechanical parameters of the pipe roof are listed in

Table 3. Mechanical parameters of advance pipe roof.

	Unit Weight (kN/m3)	Elastic Modulus (GPa)	Cohesion (MPa)	Poisson’s Ratio	Internal Friction Angle (°)
Advanced pipe roof	4840	100	0.33	0.25	41

.

4.2.2. Parameter Design and Modeling of Glass Fiber Bolts

For the B2V section type in a deep-buried segment of Tunnel 8#TA, the surrounding rock mass is fractured and exhibits poor self-stability. In addition to advance pipe roof support, this section requires reinforcement with glass fiber bolts. The glass fiber bolts are manufactured from glass fiber-reinforced thermosetting polyester resin, with a glass fiber content of not less than 60% by weight. They feature a maximum graded thread of 60 mm, an outer diameter of 60 mm, an inner diameter of 40 mm, and a thickness of 10 mm. Cement slurry is used for grouting. The technical parameters of the glass fiber bolt, such as distribution radius, quantity, and overlap length, are shown in Figure 14. A total of 55 bolts are installed in 5 rings arranged from the tunnel contour towards the center of the tunnel face. Each bolt has a length of 18 m with an overlap length of 8 m. Overflow grouting holes with a diameter of 6 mm are drilled in the front section of the glass fiber bolts, spaced at 20–30 cm intervals in a staggered pattern (see Figure 14).

In FLAC3D, cable structural elements are employed to establish the numerical model of the bolts (see Figure 15). When creating bolts using the cable element, the model is established by specifying three parameters via the by-ray method: the starting point of the bolt, the direction (point), and the bolt length. The mechanical parameters of the glass fiber bolts are listed in

Table 4. Mechanical parameters of glass fiber bolt.

	Unit Weight (kN/m3)	Elastic Modulus (GPa)	Cohesion (MPa)	Poisson’s Ratio	Internal Friction Angle (°)
Glass fiber bolt	1900	75	0.33	0.25	41

.

4.3. Mechanical Response Characteristics of Surrounding Rock Under Tunnel Face Advance Reinforcement

4.3.1. Response Characteristics of Displacement Field After Reinforcement

To analyze the pre-convergence and convergence deformation characteristics of the tunnel face under different support schemes, the vertical displacement of the vault at a characteristic cross-section 15 m deep into the model was selected for analysis. The pre-convergence and convergence deformation characteristics of the tunnel vault under different support measures are shown in Figure 16a. As shown in the figure, conditions 1, 2, and 3 all exhibit certain restraining effects on the deformation of the surrounding rock above the vault. However, due to differences in support characteristics among the conditions, the degree of restraint on vault settlement deformation varies. The final convergence value for condition 1 during tunnel excavation is 113.79 mm, with a maximum pre-convergence value of 104.91 mm. For condition 2, the final convergence value is 85.42 mm, with a maximum pre-convergence value of 76.29 mm. For condition 3, the final convergence value is 28.82 mm, with a maximum pre-convergence value of 18.15 mm. The final convergence value and pre-convergence value of condition 2 are reduced by 24.93% and 27.28%, respectively, compared with condition 1, and the final convergence value and pre-convergence value of condition 3 are reduced by 73.10% and 82.69%, respectively, compared with condition 1. Condition 2, with advance pipe roof support, exhibits a significant restraining effect on both the convergence and tunnel vault pre-convergence displacements. Condition 3, which combines bolts with pipe roof support, demonstrates a further enhanced restraining effect on these displacements. This combined support system significantly reduces the pre-convergence deformation rate of the tunnel vault and substantially decreases the final convergence displacement at the tunnel vault.

When the tunnel excavation reached 30 m, the relationship between the pre-convergence deformation of the vault surrounding rock and the distance from the tunnel face is shown in Figure 16b. As shown in the figure, under conditions 1, 2, and 3, the pre-convergence deformation of the advance core soil at the vault begins to change sharply within approximately one tunnel diameter ahead of the tunnel face, with a relatively high deformation rate. The majority of the pre-convergence displacement occurs within this zone. Under condition 3, both the pre-convergence displacement and deformation rate ahead of the tunnel face show significant reduction compared to conditions 1 and 2. This phenomenon is attributed to the spatial arching effect formed in the surrounding rock under different support schemes. Although the three support schemes (conditions 1, 2, and 3) demonstrate noticeable differences in restraining the pre-convergence deformation of the advance core soil at the vault, the combined support system of pipe roof and bolts exhibits superior effectiveness in controlling the pre-convergence deformation of the advance core soil.

To analyze the deformation patterns of tunnel face extrusion displacement under different support schemes, the horizontal extrusion displacement along the central axis of the tunnel face at a characteristic cross-section 30 m deep into the model was selected for analysis. The extrusion deformation curve along the longitudinal symmetry axis of the tunnel face is shown in Figure 17a. After tunnel excavation, due to the loss of constraining force from the surrounding rock, the stress state transitions from the original triaxial stress state to a biaxial stress state. The extrusion deformation of the tunnel face exhibits a trend of being less pronounced at the edges and more significant at the center. The maximum extrusion displacement of the tunnel face in condition 2 is 86.86% of that in condition 1, while in condition 3 it is 12.34% of that in condition 1. Therefore, through comparative analysis of the extrusion displacements under different conditions, the pipe roof support exhibits a certain restraining effect on the tunnel face extrusion displacement. However, the combined support system of advance pipe roof and glass fiber bolts demonstrates a more effective restraining effect on the tunnel face extrusion displacement.

To analyze the pre-extrusion displacement characteristics of the tunnel face under different support schemes, the extrusion displacement at the central point of the surrounding rock ahead of the tunnel face 30 m deep into the model was selected for analysis. The relationship between the extrusion deformation of the surrounding rock ahead of the tunnel face and the distance from the tunnel face is shown in Figure 17b. As shown in the figure, the majority of pre-extrusion deformation at the central point of the rock ahead of the tunnel face occurs within approximately one tunnel diameter from the tunnel face, with a relatively high deformation rate. The advance pipe roof demonstrates a certain restraining effect on the tunnel face’s center point displacement, yet its effectiveness is less pronounced compared to the combined control system of pipe roof and bolts. The combined support system not only significantly reduces the pre-extrusion deformation displacement at the central point of the surrounding rock ahead of the tunnel face, but also decreases the range of rapid deformation rate at this location.

To analyze the deformation characteristics of tunnel surrounding rock under different support schemes, when the tunnel excavation reached 40 m, the displacement at a characteristic cross-section 20 m deep into the model was selected for analysis. The time-displacement curves of the tunnel surrounding rock under different conditions are shown in Figure 18. The monitoring points for the displacement curves of the left/right haunches, vault, and arch bottom are located at the characteristic cross-section 20 m deep into the model. The horizontal displacements of the left and right haunches and the vertical displacements of the vault and arch bottom of the tunnel surrounding rock are essentially symmetrical. At approximately two tunnel diameters from the monitoring points, the surrounding rock begins to develop displacement toward the tunnel cavity. The rate of displacement change increases with proximity to the monitoring points at the characteristic cross-section. At a location approximately one tunnel diameter away from the monitoring point, the convergence rate of the surrounding rock towards the tunnel interior increases rapidly. After the tunnel face reaches the monitoring point, the displacement rate begins to gradually decrease. When the tunnel face passes the monitoring point by a distance of about one tunnel diameter, the deformation of the surrounding rock starts to gradually level off and stabilize. Compared to condition 1, the vault convergence in condition 2 is reduced by 25%, and the arch bottom uplift is reduced by 5.88%. In condition 3 relative to condition 1, the vault convergence decreases by 75%, and the arch bottom uplift decreases by 52.94%. Before the application of stabilization measures, the displacement at the vault was always greater than that at the arch bottom. After implementing advanced pipe roof support and glass fiber bolt reinforcement, the displacement at the vault becomes consistently smaller than that at the arch bottom. This indicates that the application of both methods contributes more significantly to the stability of the tunnel surrounding rock.

4.3.2. Response Characteristics of the Stress Field After Reinforcement

Taking the tunnel center point as the coordinate origin, an analysis was conducted on the variation in the major and minor principal stresses along the path directly above the tunnel. The stress path curves at the tunnel vault under different support conditions are shown in Figure 19. As can be seen from the figure, point I represents the inner boundary of the vertical pressure arch at the tunnel vault, which occurs at the point of peak principal stress. Point O denotes the outer boundary of the vertical pressure arch at the tunnel vault, where the direction of the major principal stress begins to deviate. In Figure 19c, the significant difference between the stresses at the inner and outer boundaries is due to the stress concentration phenomenon occurring in the surrounding rock near the tunnel profile. As the support conditions change, the inner and outer boundaries of the pressure arch directly above the vault also vary, meaning that the distances from these boundaries to the tunnel center point change. Compared to condition 1, the advanced pipe roof support in condition 2 has a relatively limited influence on the extent of the inner and outer boundaries of the pressure arch. In contrast, the combined support measure of advanced pipe roof and glass fiber bolts at the tunnel face in condition 3 reduces the formation range of the radial arching effect around the tunnel to an area near the tunnel profile, at a certain distance outside the contour line compared to both condition 1 and condition 2.

Figure 20 shows the stress paths of the surrounding rock at the tunnel face under different support conditions when the tunnel is excavated to 30 m. After excavation, the stress in the surrounding rock undergoes redistribution; however, due to the different support schemes, the stress states in the surrounding rock near the tunnel face exhibit varying conditions. From condition 1 to condition 3, the extent of stress variation around the tunnel contour gradually decreases as the support measures are enhanced. In both the area above the vault and the region below the arch bottom, the range over which the direction of the major principal stress deflects is also progressively reduced. Furthermore, the distance from the tunnel face to the zone of virgin rock stress distribution ahead of the tunnel gradually shortens, meaning that the extent over which the major principal stress deflects decreases. The range of principal stress deflection in the upper stratum of the tunnel exceeds that in the lower stratum, indicating that excavation activities are more likely to trigger local instability in the upper formation. When stress deflection forms an arching effect, the overburden load is transferred through this stress arch to certain areas ahead of and behind the tunnel face. Therefore, to prevent instability of the surrounding rock, reinforcement measures should be prioritized within a specific range both ahead of and behind the tunnel face.

The stress path of the surrounding rock is shown in Figure 21. The monitoring points for the stress curves of the left and right haunches, as well as the vault and arch bottom, are located at the characteristic cross-section of the model at the longitudinal position of 20 m. The rock mass was in a stable state before tunnel excavation. The excavation process induced stress redistribution in the surrounding rock. After excavation, the stress variation curves at the left and right haunches of the monitoring cross-section are essentially identical, while the stress curves at the vault and arch bottom show consistent trends with slight differences in magnitude. The stress in the surrounding rock initially increases slightly during tunnel excavation. As the tunnel face approaches the 20 m mark, the stress gradually decreases. After the monitoring cross-section is excavated, the stress increases again and eventually tends to converge. When advanced support measures are implemented to reinforce the surrounding rock, the stresses under condition 1, condition 2, and condition 3 gradually increase after excavation. This occurs because the enhanced stabilization measures promote the formation of a pressure arch around the tunnel, which helps control deformation of the surrounding rock and reduces stress release to some extent.

4.3.3. Response Characteristics of the Plastic Zone After Reinforcement

The distribution of the plastic zone in the surrounding rock under different conditions when the tunnel is excavated to 40 m is shown in Figure 22. It could be observed that the advanced pipe roof in conditions 2 and 3 effectively suppresses the development of the plastic zone at the vault. Furthermore, in condition 3, the development of the plastic zone in the advance core soil ahead of the tunnel face is also well controlled. Moreover, the formation and evolution characteristics of the plastic zone during tunnel excavation can be clearly observed in the figure. Plastic deformation occurs both around the tunnel periphery and ahead of the tunnel face. The deformation zones in the tunnel cross-section are primarily dominated by shear failure, while the surrounding rock ahead of the face mainly undergoes tensile-shear failure. The concentrated distribution of the plastic zone indicates that the stability of the surrounding rock in these areas is relatively vulnerable, requiring corresponding support measures to prevent failure and collapse.

The built-in FISH language in the software was used to statistically analyze the plastic zone volumes under the three support conditions. The obtained shear failure volumes for conditions 1, 2, and 3 are 22,558.65 m³, 14,863.63 m³, and 3657.69 m³, respectively, while the tensile failure volumes are 287.59 m³, 540.99 m³, and 40.55 m³, respectively. It can be observed that the volume of shear failure in the tunnel gradually decreases as the support measures are enhanced. However, the volume of tensile failure in the plastic zone undergoes a significant mutation change in condition 3. This is primarily due to the additional reinforcement provided by bolts in the advance core soil area, which was implemented in condition 3 on the basis of conditions 1 and 2. This reinforcement effectively restrains further relaxation and deformation of the soil, thereby reducing the volume of tensile failure in the plastic zone. These results demonstrate that the application of face bolts plays a significant role in controlling the expansion of the plastic zone during tunnel excavation and ensuring tunnel stability.

5. Result Verification

To verify the rationality and reliability of the numerical simulation results in this paper, the field measured data and simulation results are compared and analyzed. The on-site monitoring section is selected at DK11 + 125 of 8# tunnel in F3 section of E60 expressway in Georgia. Before the construction of the secondary lining of the tunnel, TCR702 total station, JSS30A30 convergence meter, CCD binocular camera and other equipment were used to monitor the deformation of the section (Figure 23a). The stability type of the surrounding rock of this section is B, and it is far from the geological structure of the fault. The geological conditions are relatively stable, and it is suitable as a comparison object to verify the simulation results. However, limited by the construction site conditions and monitoring technical means, this study failed to obtain the time-history monitoring data of the whole process of surrounding rock deformation. It can only obtain the final displacement values of several key parts (vault, arch bottom and face center) under the condition of combined support (condition 3) after the tunnel excavation is stable (Figure 23b). Although the data set is limited, these final displacements are one of the most direct and important indicators for evaluating the support effect. To verify the model, the final displacement value obtained by numerical simulation is compared with the field monitoring results (see

Table 5. Data comparison between numerical simulation and field measures.

Different Parts	Data of the Displacement (mm)
	Vault	Arch Bottom	Face Center
Field monitoring	36.01	43.68	63.8
Simulation	30.61	39.41	55.17

).

Based The comparative analysis shows that the numerical simulation results are in good agreement with the field monitoring data in terms of change trend and numerical magnitude. Specifically, both the simulation and the measured results show that the uplift of the arch bottom > the extrusion of the working face > the settlement of the vault. This law accurately reflects the deformation characteristics of the tunnel under specific geological and supporting conditions. At the same time, it can be found that the simulated values are generally slightly smaller than the measured values, and the relative errors of the vault, arch bottom and tunnel face are 15.0%, 9.8%, and 13.5%, respectively. This deviation is within the reasonable acceptable range of numerical analysis of geotechnical engineering. The main reasons for the deviation can be summarized as follows: Firstly, the actual rock mass has significant heterogeneity and anisotropy, and contains complex geological structures such as joints and fissures. However, the numerical model simplifies it into homogeneous isotropic material, which will inevitably bring some differences. Secondly, the possible groundwater softening effect and the non-ideal complete common deformation relationship between the supporting structure and the surrounding rock are not fully reflected in the model. Despite these simplifications, the simulation results are highly consistent with the limited field key data on the core indicators, which fully proves that the three-dimensional numerical model established in this study, the selected constitutive relationship and parameters have reasonable accuracy and reliability. Therefore, based on the model, the stability evaluation of the unsupported state, the comparative analysis of different support schemes and the conclusions about the synergistic mechanism of combined support are credible, which can provide valuable reference for similar projects.

6. Discussion

As mentioned previously, the numerical simulation results of this study, within the theoretical framework of ADECO-RS, provide valuable insights into the effectiveness of pre-reinforcement for tunnel face stability in weak rock masses. A systematic comparison of different support schemes clearly demonstrates the advantages of pre-support in controlling deformation, optimizing the pressure arching effect, and reducing the plastic zone, with a quantitative analysis of the synergistic effects of the combined pre-reinforcement system. These findings offer significant reference value for similar engineering conditions. However, the numerical model in this study intentionally adopted isotropic and homogeneous rock mass assumptions, and did not explicitly consider geological structures (such as faults) or groundwater effects. This simplification was based on a specific research objective: to isolate and clearly reveal the fundamental reinforcement mechanisms and synergistic effects of the combined system of an advanced pipe roof and glass fiber bolts under controlled conditions. Introducing full geological complexity at this stage would make it difficult to distinguish whether the observed mechanical responses (such as deformation control and pressure arch evolution) were due to the support measures or the combined result of rock mass heterogeneity and hydro-mechanical coupling effects, thereby obscuring the core mechanisms.

6.1. Limitations of the Numerical Model

• As the numerical model of this study adopts the assumption of homogenization, which fails to fully consider the rapid change in lithology and weak structural plane in the field, and ignores the inherent spatial variability and uncertainty of geotechnical parameters [23]. This may lead to that the deformation field predicted by the model is more uniform and less serious than the actual situation, which cannot fully reflect the significant asymmetry and deformation localization characteristics that may occur in heterogeneous rock masses. Secondly, the model depends on the deterministic mean values of parameters such as elastic modulus (E), cohesion (c), internal friction angle (φ) and initial stress state (K₀) [23]. The natural variability and uncertainty of these input parameters means that our results represent only a specific deterministic result based on the best estimate. For example, a lower strength parameter (c, φ) will undoubtedly lead to a larger plastic zone and more severe deformation, while a lower deformation modulus (E) will increase the displacement.
• In addition, the lack of consideration of groundwater in the model is a key limitation. As shown in Figure 1e, the phenomenon of water and mud gushing in tunnel construction will significantly weaken the shear strength of surrounding rock by reducing the effective stress of rock mass and accelerating softening and weathering under real conditions, thus further aggravating the deformation and expanding the range of plastic zone. This makes the current conclusion more suitable for ideal conditions under drainage or drying conditions. In particular, it should be pointed out that the reinforcement mechanism of glass fiber bolts concerned in this paper is mainly aimed at the stability of broken rock mass under the redistribution of excavation stress, but does not simulate large active faults with significant shear displacement. Moreover, due to its brittleness and limited shear resistance, it is difficult to resist the shear deformation caused by fault dislocation. Its core function is to strengthen the anti-extrusion ability of the core soil in front of the tunnel, rather than as a shear member crossing the continuous active fracture surface.

The simplified factors will unavoidably cause some differences between numerical simulation results and actual engineering responses. To improve prediction accuracy and applicability, future research will build on the findings of this work and focus on two aspects: (1) In terms of model development, using advanced models that can better describe the softening and expansion behaviors of rock and soil after the peak, to better understand the gradual failure process of surrounding rock. (2) Regarding external conditions, establishing fluid–mechanics coupled models and explicitly including geological features like faults, to accurately assess the combined effects of groundwater and structural shearing on the performance of support systems.

6.2. Engineering Value

The numerical analysis results of this study provide clear practical guidance for tunnel engineering design and construction under similar geological conditions. Based on the above conclusions, the key engineering enlightenment is that for tunnels passing through weak broken strata and with poor self-stability of surrounding rock (corresponding to tape B or tape C in ADECO-RS classification), active advance support strategy must be adopted. Specifically, when monitoring or prediction shows that there is a significant risk of pre-convergence (e.g., more than 100 mm) or face extrusion deformation (e.g., more than 300 mm) in the surrounding rock, it is strongly recommended to use the combined support scheme of advanced pipe roof and glass fiber bolts rather than a single support measure. The combined scheme can form an effective transverse arch shell and longitudinal beam bearing system through the pipe roof, and use the glass fiber bolt to enhance the strength and integrity of the core soil in front, so as to synergistically improve the deformation control ability and significantly inhibit the expansion of the plastic zone. In terms of parameter design, this study confirms that the deformation is mainly concentrated in the range of about one time of the tunnel diameter in front of the tunnel face. Therefore, in the design and construction, it is necessary to ensure that the advanced support system (including pipe roof and bolt) has sufficient longitudinal length to completely cover and exceed the aforementioned main deformation affected area. At the same time, it is necessary to ensure that there is a sufficient overlap length between the front and rear support rings to form a continuous and complete reinforcement area in space, so as to avoid the formation of weak links at the end of the support. Finally, all theoretical analysis and numerical conclusions should be verified through a rigorous on-site monitoring network during construction to form a dynamic closed-loop of ‘design-construction-monitoring-feedback’, so as to achieve precise control of tunnel safety construction.

7. Conclusions

Based on the No. 8# TA Tunnel project in the E60–F3 section of Georgia, this study established a three-dimensional model using the finite difference method to simulate full-face excavation of a large cross-section tunnel. The stability of the tunnel face was evaluated by analyzing the multi-physical response of the surrounding rock. Three support schemes were compared by simulating the pre-convergence/convergence displacement and the extrusion displacement of the tunnel face during excavation. Changes in the stress field and the distribution of the plastic zone were also investigated to evaluate the effectiveness of advanced support. The main findings are as follows:

• Stability classification as type B: The unsupported analysis is crucial for classification. It reveals critical deformations (face extrusion of 382.8 mm and vault settlement of 112.4 mm) with a pressure arch forming 17.5–22 m from the tunnel. The failure mode is complex (tensile shear at the face and arch bottom, shear at the vault). These responses definitively identify the surrounding rock as type B (short-term stable), which means that advanced support must be used to prevent instability, thus verifying the necessity of the support scheme discussed in this study.
• Quantitative superiority of combined support: The combined support (pipe roof + glass fiber bolts) achieves decisive deformation control. Compared with the primary support, it has achieved excellent reduction effect: final convergence is reduced by 73.10%, pre-convergence is reduced by 82.69%, and extrusion of the face is reduced by 87.7%. The pipe roof alone offers moderate improvement. This sharp contrast shows that the combined support has superior synergistic effect in ensuring the stability of the face and controlling the pre-convergence.
• Fundamental mechanical improvement: The combined support fundamentally enhances the rock mass state. It dramatically contracts the pressure arch boundaries to 6–12.5 m, promoting a more efficient load-bearing structure. Crucially, it reduces the shear failure volume by about 84% (to 3658 m³) and significantly reduces the tensile failure. This highlights the key role of glass fiber anchors in strengthening the core soil, controlling the expansion of the plastic zone and ensuring overall stability, and reveals an important mechanism behind the success of the support system.