Mechanical Effects of Lining Thinning in Shallow Four-Track High-Speed Railway Tunnels: Field Monitoring and Numerical Analysis

Abstract

Lining thinning is a common defect in railway tunnels; however, its impact on shallow four-track high-speed railway (HSR) tunnels—particularly on rock pressure—remains poorly understood. This study investigates the influence of lining thinning on the genuine pressure (also referred to as “deformation pressure”) of such tunnels through field investigation, long-term monitoring, and numerical simulation. Firstly, a lining thinning survey was conducted across ten tunnels, and the statistical distribution of defect parameters was analyzed. Second, over 180 days of field monitoring were carried out in China’s first four-track HSR tunnel (XBS Tunnel) to evaluate the rock pressure state. Third, a three-dimensional numerical model was developed to evaluate the effects of lining thinning and structural degradation on the principal stress, deformation, and genuine pressure of shallow four-track HSR tunnels. The results indicate that thinning defects are widespread, with 38.64% occurring at the vault and over 84% having minimum thicknesses below 0.26 m. The actual rock pressure in the XBS Tunnel was significantly lower than theoretical predictions, and the tunnel primarily experienced genuine pressure rather than loosening pressure during construction, with the secondary lining serving as a safety reserve. Lining thinning leads to stress redistribution and concentration, weakens structural stiffness, and increases the likelihood of damage. It also induces a transition from genuine pressure to loosening pressure, a process that is further accelerated by surrounding rock and lining degradation. The findings provide important insights for the evaluation, design, and long-term maintenance of large-span HSR tunnels.

1. Introduction

High-speed railway (HSR) tunnels play a critical role in modern transportation infrastructure, enabling rapid, large-volume, and energy-efficient mobility across vast regions. As operational demands increase, the construction of super-large-span tunnels—particularly single-bore four-track configurations—has become necessary to accommodate growing traffic and limited underground space [1,2]. However, railway tunnels are highly susceptible to both construction-related defects and operational degradation over time. Among these, lining thinning and voids have been frequently observed [3,4,5], as shown in Figure 1. Given the structural scale and functional importance of four-track HSR tunnels, accurately assessing the mechanical implications of thinning defects has become a critical technical challenge—yet remains insufficiently addressed in current engineering practice.

In recent years, numerous studies have focused on the structural performance of tunnel linings under various defects, including voids, cracks, and material degradation. These studies have primarily focused on three aspects: defect detection, impact assessment, and remediation strategies [6]. In the field of defect detection, Xue and Li [7] proposed an automatic classification and detection method for shield tunnel lining defects using deep learning. Gao et al. [8] developed a deep learning-based FCN-RCNN model for the simultaneous detection of multiple defects in subway tunnels. To identify surface defects in tunnel linings, Li et al. [9] designed the MTSIS system for efficient defect acquisition and analysis. In the area of defect impact assessment and remediation, Qin et al. [10] investigated the influence of insufficient vault thickness in a tunnel crossing the Yellow River and proposed remediation measures involving concrete backfilling and steel plate reinforcement. Using the cohesive zone model, Liu et al. [11] analyzed failure mechanisms induced by concrete cracking and lining thickness reduction. Chen et al. [12] studied the effects of voids and insufficient lining thickness on stress distribution, internal forces, and safety factors in tunnels. Han et al. [13] provided a comprehensive review of current defect detection techniques and, based on the analytic hierarchy process (AHP), identified key causes of voids and thinning, recommending fiber-reinforced plastic (FRP) as a reinforcement solution. Additionally, Chen et al. [4] examined the mechanical behavior and failure modes of tunnel linings under the combined influence of voids and thinning through both numerical simulation and physical model testing. However, several key limitations remain. For example, in the area of impact assessment, the majority of numerical and theoretical investigations have focused on single- or double-track tunnels, with almost no studies addressing shallow-buried four-track HSR tunnels—a structure with distinct geometric and mechanical characteristics [14]. Most importantly, existing studies are predominantly based on the loosening pressure theory, which assumes that the tunnel lining bears the full weight of the overlying rock mass. However, most modern HSR tunnels constructed using the mining method are guided by the principles of the New Austrian Tunneling Method (NATM) [15], employing a composite lining system that consists of pre-support, primary support, and secondary lining as the core structural components [16]. The functions of pre-support and primary support include the following: (1) acting as a flexible structure that remains in close contact with the surrounding rock to control deformation during excavation and prevent rockfall, weathering, extrusion, and swelling; (2) coordinating deformation with the surrounding rock to mobilize its self-supporting capacity, thereby forming an integrated load-bearing system; (3) locally or globally improving the mechanical properties of the rock mass through the application of shotcrete, rock bolts, and grouting; and (4) shielding the rock surface to prevent further weathering. In tunnels with conventional cross-sections and typical geological conditions, the secondary lining is generally regarded as a structural safety reserve. Accordingly, with the widespread implementation of NATM-based anchor–shotcrete systems and modern monitoring technologies, the dominant form of surrounding rock pressure in deep-buried mountain railway tunnels has theoretically transitioned from loosening pressure to genuine pressure(also referred to as “deformation pressure”) [17,18].

Despite this shift, the influence of lining thinning on genuine pressure has received limited attention in the existing literature, leaving a critical gap in both theoretical understanding and engineering practice. Furthermore, most previous studies have primarily focused on conventional single- or double-track tunnels. In recent years, however, the need for lane merging, track expansion, and increased transport capacity has led to the construction of super-large cross-section tunnels in both highway and railway systems. Four-track, ultra-large-span high-speed railway (HSR) tunnels have begun to emerge in engineering practice. These tunnels typically adopt sequential excavation methods [19], which involve complex procedures, large volumes of concrete lining, and considerable challenges in construction quality control. As a result, defects such as lining thinning and voids become almost inevitable. Without timely mitigation, such defects—exacerbated by long-term structural degradation—may lead to serious damage to the tunnel lining. Therefore, further research is urgently needed to investigate the mechanical behavior and safety implications of lining defects in ultra-large-span tunnels.

This study investigates the influence of lining thinning on the genuine pressure in shallow four-track HSR tunnels through a combination of field investigation, long-term monitoring, numerical simulation, and theoretical analysis. First, a lining thinning survey was conducted across ten tunnels using a tunnel inspection vehicle, and the statistical distribution of defect parameters was analyzed. Second, over 180 days of field monitoring were carried out in China’s first four-track HSR tunnel. The monitoring data were used to assess the surrounding rock pressure state of the XBS Tunnel. Third, a three-dimensional numerical model established by using Abaqus Software (Version 2022) was developed to evaluate the effects of lining thinning and structural degradation on the principal stress, deformation, and genuine pressure of shallow four-track HSR tunnels.

2. Field Investigation on Lining Thinning Defects in Railway Tunnels

2.1. Investigation of Thinning Defects

A comprehensive investigation of lining thinning defects was conducted across ten railway tunnels using a tunnel inspection vehicle covering a total length of 40,187 m. The inspection vehicle was equipped with three ground-penetrating radar (GPR) antennas and operated at a speed of 5 km/h. The nine radar antenna units operated at frequencies of 400 MHz + 900 MHz, with an effective detection depth of approximately 1.0 m. The inspection scene is illustrated in Figure 2. The inspection focused on identifying the following key characteristics of thinning defects: the tunnel mileage at which the defect occurred, the defect location within the tunnel cross-section (P_j), the longitudinal extent of the thinning defect (L_j), the minimum thickness of the lining after thinning (H_min), and the average thickness of the lining after thinning (H_avg). Core drilling was performed in selected sections to verify the accuracy of the radar detection results, as shown in Figure 2c. A typical radar image indicating insufficient lining is presented in Figure 3.

2.2. Statistical Characteristics of Thinning Defects

Seven survey lines were established for the defect investigation, as shown in Figure 4a. The spatial parameters of the thinning defects—namely, the defect position within the cross-section (P_j), defect length (L_j), average thickness (H_avg), minimum thickness (H_min), and the thinning–thickness ratio (R_j)—are presented in Figure 4b–f. Here, the thinning–thickness ratio R_j is defined as the ratio of the average lining thickness (H_avg) after thinning to the designed lining thickness (H_des). Thinning defects were most frequently observed at the vault (corresponding to survey line 4), accounting for 38.64% of all cases. The remaining defects were more evenly distributed, with proportions ranging from 6.82% to 15.91%. In terms of defect length, the most common range was 4–6 m (43.18%), followed by 6–8 m (25.00%). The maximum observed defect length was 17.0 m. Regarding the average thickness, 90% of thinning defects fell within the range of 0.175 to 0.325 m, with the minimum average thickness recorded as 0.15 m. For minimum thickness values, 84.11% were between 0.14 and 0.26 m, with the lowest value observed being 0.09 m. The thinning–thickness ratio R_j was most commonly in the range of 0.275 to 0.325 (40.9%), with the lowest recorded value being 0.25. These findings indicate the presence of severe thinning defects in the surveyed railway tunnels.

3. Field Monitoring of Shallow Four-Track HSR Tunnels

3.1. Monitoring of a Shallow Four-Track HSR Tunnel

The XBS Tunnel of the Hangzhou–Shaoxing–Taizhou (HST) railway consists of Tunnel 1 and Tunnel 2, with lengths of 166 m and 430 m, respectively. Considering factors such as track merging near stations, the tunnel was designed as a single-bore four-track tunnel—the first of its kind in China. With a span ranging from 26.3 m to 27 m and a burial depth of 7–53 m, it represents a rare example of a super-large-span tunnel. The tunnel portal section passes through silty clay and tuff with well-developed joints and fractures, resulting in extremely fragmented rock masses classified as Grade V rock [16]. The main tunnel alignment traverses tuff formations, predominantly of Grades IV and V. Based on engineering analogy and geological conditions, a composite support system of “advanced support + double-layer primary support + secondary lining” was adopted for the XBS Tunnel, as illustrated in Figure 5. The construction site is shown in Figure 6. The XBS four-track HSR tunnel has been in operation since January 2022 and is currently performing well.

Field monitoring of the XBS four-track HSR tunnel was conducted over a period of more than eight months from February to November 2019. The monitoring program included observations of rock deformation, deformation of the support structure, contact pressure between the rock and the primary support, stresses in steel arch frames, stresses in lattice girders, concrete stress within the primary support, and concrete stress in the temporary support. A selection of monitoring instruments is summarized in

Table 1. Selected monitoring instruments used in the XBS Tunnel.

Monitoring Item	Instrument Type	Measurement Range	Accuracy	Quantity Per Section
Contact pressure	XYJ-2 pressure cell	1 MPa	<0.5% F.S.	18 units
Stress in steel arches of temporary support	Surface strain gauge (steel)	3000 με	Compression: <0.5% F.S. Tension: <0.1% F.S.	12 units
Stress in lattice girders of primary support	XZ-B type φ22 rebar stress gauge	200 kN	±0.1% F.S.	36 units
Shotcrete stress in primary support	XJH-2 strain gauge	2000 με	±0.1% F.S.	36 units
Concrete strain in temporary support	XJH-2 strain gauge	2000 με	±0.1% F.S.	18 units
Axial force of anchor bolts	Anchor bolt load cell	100 kN	±0.1% F.S.	7 sets

. The layout of monitoring points is shown in Figure 7, and the on-site monitoring setup is presented in Figure 8.

3.2. Analysis of Field Monitoring Data

The distribution of rock pressure in the XBS Tunnel is illustrated in Figure 9, using sections DK215+105 and DK215+142 as representative examples. Based on the rock pressure curves and deformation profiles, the rock pressure and rock mass deformation under different burial depths and rock mass grades were extracted and summarized in

Table 2. Comparison of measured data.

Burial Depth	Rock Grade	Sample Size	Field Monitoring (kPa)	Code-Based Values (kPa)	Protodyakonov’s Theory (kPa)	Full Overburden (kPa)	Settlement (mm)	Convergence (mm)
H ≤ 1.0 B	V	5	48.35~131.26	115.12~328.16	-	123.34~420.21	39.7	18.8
	IV	13	110.01~196.16	189.09~339.73	-	211.06~440.26	49.5	28.9
1.0~1.5 B	V	5	126.60~189.01	409.40~455.74	-	556.92~705.28	47.8	22.4
	IV	5	126.30~157.65	230.43~257.93	328.60~367.81	-	15.4	7.7
	III	6	116.19~124.72	132.12~140.23	184.93~196.29	-	18.1	10.1
1.5~2.5 B	IV	8	124.24~156.92	229.19~254.21	326.83~360.74	-	24.6	16.2
	III	14	110.96~126.32	130.26~139.61	182.33~195.42	-	22.4	12.7

.

Field monitoring data from the shallow-buried four-track HSR tunnel indicate that the measured rock pressures were significantly lower than those calculated using the specifications outlined in the Chinese Tunnel Code [16], Protodyakonov’s theory [20], and the full overburden weight method. Specifically, for Grade V rock, the measured pressures accounted for only 30.92% to 42.00% of the theoretical values derived from empirical formulas; for Grade IV, the values ranged from 54.21% to 61.73%; and for Grade III, they reached 85.18% to 90.48%. Additionally, the maximum vault settlement and horizontal convergence measured during monitoring were 49.5 mm and 28.9 mm, respectively—well below the design allowances of 150 mm and 200 mm. Similarly, the internal forces in the support system—including the concrete lining, steel arches, and anchor bolts—remained within acceptable limits. No excessive deformation or structural damage was observed in either the rock mass or the supporting elements. Furthermore, no signs of loosening or collapse were detected in either shallow or deep rock zones of the XBS Tunnel. The tunnel primarily passes through tuff, a relatively intact, water-free, and geologically stable rock type. The adopted support scheme—consisting of “advance support + dual-layer primary support + secondary lining”—was effective in enhancing the self-bearing capacity of the rock and promoting cooperative deformation between the rock mass and support structure. As of this writing, the tunnel remains in a stable and safe operational state. Based on the rock pressure mechanism, field monitoring data, and operational performance, it can be concluded that the XBS Tunnel experienced a genuine pressure state characterized by cooperative deformation and shared load-bearing between the rock and the support system during the construction phase.

3.3. Mechanism of Genuine Pressure

Rock pressure refers to the force exerted on the surrounding rock and support structures during underground excavation, which may lead to deformation or structural failure. Loosening pressure is defined as the load directly imposed on the support structure by detached or collapsed rock masses under the influence of gravity. Genuine pressure occurs when the deformation of the surrounding rock is restrained by closely fitting support systems, leading to a coordinated deformation and load-sharing mechanism between the rock mass and the support. In this context, the rock exerts contact pressure on the supporting structure. With the advancement of tunneling techniques, construction philosophies, support systems, and modern monitoring technologies, the concept of rock pressure in conventional railway tunnels has evolved—from the traditional loosening pressure to what is now termed genuine pressure. Based on field monitoring during the construction of the Zhengzhou–Wanzhou HSR tunnel, Wang et al. [21] determined that the rock pressure in mechanically excavated tunnels corresponded to loosening pressure. The continuous evolution of tunnel engineering practices is expected to reinforce this transformation. It can be concluded that the primary direction in current tunnel engineering is to control the rock pressure within the genuine pressure stage, as illustrated in Figure 10. Based on the compatibility between the tunnel lining and rock, as well as the rock mass quality, this study proposes a conceptual diagram of tunnel bearing states (Figure 11). Assuming the tunnel is initially in a genuine pressure state (point A) upon completion, then if the rock deteriorates, the tunnel may shift to a loosening pressure state (point B). If the lining itself deteriorates, the bearing state may evolve to point C. When both the lining and rock undergo simultaneous degradation, the tunnel’s load-bearing state may reach point D.

4. Influence of Lining Thinning on Rock Pressure in Four-Track HSR Tunnels

4.1. Numerical Model Development

A three-dimensional numerical model of an HSR tunnel with lining thinning and thinning defects was established using Abaqus (Version 2022). The model dimensions were 92.7 m in length, 50 m in width, and 100 m in height, as shown in Figure 12a. The surrounding rock mass was assumed to be homogeneous and isotropic, which is a common simplification in tunnel simulations to facilitate computational feasibility and ensure modeling clarity. The rock behavior was simulated using the Mohr–Coulomb elastoplastic constitutive model, while the primary support and secondary lining were modeled as linear elastic materials. Surface-to-surface contact interactions were defined between the primary support, the secondary lining, and the rock. Boundary conditions included normal constraints on all side boundaries and fixed constraints at the model base.

To facilitate the analysis of different scenarios, thinning s at various locations were assigned reference numbers, as shown in Figure 12b. The region directly influenced by the thinning was defined as the thinning-affected zone, while the surrounding adjacent region was referred to as the active thinning zone, as illustrated in Figure 12c. The loading and simulation conditions used in the numerical study are listed in

Table 3. Numerical simulation conditions.

Condition	Description	Control Parameters	Rock Grade
1	No defect	-	III, IV, V
2	Lining thinning	Pj, Larcj, Lj, Rj	III, IV, V
3	Lining degradation + thinning	Pj, Larcj, Lj, Rj, t	III, IV, V
4	Rock degradation + thinning	Pj, Larcj, Lj, Rj, t	III, IV, V
5	Rock degradation + lining degradation + thinning	Pj, Larcj, Lj, Rj, t	III, IV, V

.

While the assumptions of homogeneity and isotropy may not fully capture the inherent heterogeneity and anisotropic behavior of natural rock masses, they remain the most widely adopted simplifications in current engineering practice—particularly when site-specific anisotropic parameters are unavailable. Within the context of this study, which focuses on comparative analysis and parametric evaluation at the engineering scale, the use of these assumptions is considered reasonable and appropriate. In future work, the influence of spatial variability and heterogeneity in rock mass properties will be further considered to improve the accuracy and realism of the simulation results.

The thinning defect of the tunnel lining was modeled as a sector-shaped body characterized by three key parameters: the thinning length, the thinning–thickness ratio, and the thinning arc length. The mechanical parameters of the tuff were obtained through field sampling and laboratory testing conducted at the XBS Tunnel site. The method used to determine the rock mass parameters is illustrated in Figure 13. The parameters adopted for the numerical simulation are summarized in

Table 4. Physical and mechanical calculation parameters.

Materials	Elasticity Modulus (GPa)	Density (kg/m3)	Poisson’s Ratio	Cohesive Force (kPa)	Internal Friction Angle (°)
Primary support	27.33	2550	0.2	-	-
Temporary support	29.65	2500	0.2	-	-
Class III rock	1.72	2300	0.26	2232	24.3
Class IV rock	0.73	2200	0.30	1372	16.9
Class V rock	0.35	2000	0.33	245	9.1

.

4.2. Stress Analysis Results

The stress distribution in the tunnel lining under the influence of thinning defects was analyzed for a shallow four-track HSR tunnel. The representative results are shown in Figure 14, based on engineering conditions of 34 m burial depth and Grade IV rock. Whether viewed from the outer or inner surface, the presence of thinning defects leads to a significant stress redistribution in both the defect-affected zone and the adjacent active zone of the shallow four-track HSR tunnel. Pronounced stress concentration is observed in the lining structure. Moreover, based on the area of the stress concentration zone, it is evident that the thinning defect has a greater impact on the inner side of the lining than on the outer side. This is attributed to the modeling assumption that the thinning defect occurs on the inner surface of the lining in the numerical simulation. Under the thinning defect conditions illustrated in Figure 14a,c,d, the maximum principal stress on the outer side of the lining reaches its peak within the thinning-affected zone, while on the inner side, the maximum stress appears in the active thinning zone. The peak values of maximum principal stress under these three scenarios are 2.066 MPa, 5.015 MPa, and 4.496 MPa, respectively. In contrast, under the scenario shown in Figure 14b, the highest maximum principal stress occurs on the inner side of the vault lining, reaching a peak value of 6.674 MPa.

Based on the stress contour maps of the tunnel lining, the distribution curves of the principal stress along the axial direction of the thinning defect were extracted. The influence of different thinning defect parameters—including position (P_j), longitudinal extent (L_j), arc length (L^arc_j), and thinning–thickness ratio (R_j)—on the maximum principal stress is illustrated in Figure 15. The analysis is based on the engineering conditions of 34 m burial depth and Grade IV rock.

As shown in Figure 15a, the presence of thinning defects causes the distribution of the maximum principal stress along the defect axis to exhibit a W-shaped pattern at positions P_j = 1 to 6 and an M-shaped pattern at P_j = 7, 8, and 9. The peak maximum principal stress reaches 3.89 MPa at P_j = 6, while the lowest value of 0.54 MPa occurs at P_j = 5. When the defect is located at P_j = 7 or beyond, the variation in maximum principal stress is minimal, and the curve approaches a nearly horizontal line. Figure 15b shows that the peak maximum principal stress of the lining affected by thinning defects typically appears near the boundary of the active thinning zone, whereas the minimum stress is observed within the thinning-affected zone. As the longitudinal extent L_j increases, the W-shaped stress distribution widens. However, the maximum value of principal stress does not increase correspondingly; for instance, when L_j = 8 m, the peak stress reaches 5.53 MPa, while increasing L_j to 24 m causes a reduction in the peak stress to 3.73 MPa. According to Figure 15c, when the thinning arc length L^arc_j is 4 m, the maximum principal stress is 1.13 MPa. When L^arc_j increases to 8 m and 16 m, the maximum stress rises sharply to 5.01 MPa. However, further increases to 32 m and 40 m result in decreased σ_max values of 4.31 MPa and 4.04 MPa, respectively. As illustrated in Figure 15d, when the thinning–thickness ratio R_j < is 0.75, the W-shaped stress distribution is nearly flat, and the peak principal stress is 2.18 MPa. When R_j is reduced to 0.50, σ_max increases to 3.81 MPa. A further reduction to R_j = 0.25 causes a more pronounced W-shape with a greater vertical variation, although σ_max decreases to 1.84 MPa.

According to the railway tunnel design code [16], the flexural tensile strength of C30 concrete is 0.55 MPa. The simulation results show that the maximum principal stresses caused by thinning defects significantly exceed this allowable stress, indicating that tensile failure has likely occurred and cracks have already developed in the lining. The load-bearing capacity of the active thinning zone is weakened, and the load originally carried by the thinned lining is redistributed to the adjacent lining in the thinning-affected zone, resulting in stress concentrations in both zones. Furthermore, due to the presence of thinning defects, the load that would otherwise be directly transmitted to the secondary lining is instead redistributed through the adjacent primary lining. This increases the load transfer demand within the thinning-affected zone, leading to stress concentration, especially on the inner side of the lining within this zone.

4.3. Deformation Analysis Results

The stress distribution in the tunnel lining under the influence of thinning defects is shown in Figure 16, with the defect parameters identical to those in Figure 14. Thinning defects lead to significant settlement within the active thinning zone of the tunnel lining. The deformation contour plots reveal distinct elliptical or rounded-rectangle-shaped subsidence zones in the vicinity of the thinning region. Thinning defects cause pronounced vertical settlement within the active thinning zone of the tunnel lining. The deformation contour maps exhibit distinct elliptical or rounded-rectangular subsidence zones surrounding the thinning region. Under the thinning condition shown in Figure 16a, the maximum vertical settlement within the active thinning zone reaches 18.83 mm. For the case in Figure 16b, the peak settlement is also located within the active thinning zone, reaching 31.27 mm. In Figure 16c, the maximum settlement of 20.54 mm appears at the center of the thinning defect located at the vault. Similarly, in Figure 16d, the settlement peaks at the center of the active thinning zone, reaching 20.68 mm.

The reduction in lining thickness results in a localized decrease in structural stiffness, weakening the lining’s resistance to deformation under external loads. This leads to concentrated settlement and the formation of a “deformation basin”, which may become a potential risk zone for crack initiation or lining failure. Such localized deformation may further induce cracking, spalling, or structural instability. These risks are especially critical under repeated operational loads, where cumulative effects can accelerate damage progression.

Based on the deformation contour maps, the vertical settlement distribution curves of the lining along the axial direction of the thinning defect were extracted. The influence of different thinning defect parameters—namely, P_j, L_j, L^arc_j, and R_j—on the vertical settlement of the lining is illustrated in Figure 17. The analysis is based on engineering conditions with a burial depth of 34 m and Grade IV rock.

As shown in Figure 17a, when the thinning defect is located at P_j = 1 to 6, the tunnel lining exhibits a settlement trough. In contrast, when P_j = 7, 8, or 9, the lining shows a heaving deformation pattern. Figure 17b reveals that when the thinning length L_j is 4, 8, or 12 m, the settlement within the active thinning zone exhibits a W-shaped distribution. When L_j increases to 16, 20, or 24 m, the settlement profile evolves into a “W nested in M” shape. The maximum settlements for L_j = 4, 8, and 12 m are 21.03 mm, 20.37 mm, and 20.86 mm, respectively. For L_j = 16, 20, and 24 m, the corresponding maximum settlements are 21.26 mm, 20.65 mm, and 20.77 mm. With the exception of L_j = 20 m, the peak settlements consistently occur at the edges of the thinning zone, within the thinning-affected zone. When L_j = 20 m, the maximum settlement appears at the center of the active thinning zone. As shown in Figure 17c, when the thinning arc length L^arc_j is 4 m, the settlement distribution takes the form of a simple trough. For L^arc_j = 8, 16, 32, and 40 m, the settlement exhibits a W-shaped pattern, with the peak settlement occurring at the base of the trough or W shape. The maximum settlement for L^arc_j = 4 m is 19.93 mm. As L^arc_j increases, the peak settlement also increases, reaching 20.82 mm when L^arc_j = 40 m. Figure 17d shows that when the thinning–thickness ratio R_j is 0.75, the settlement profile is trough-shaped with a peak value of 23.70 mm. When R_j is 0.50, the settlement in the active thinning zone displays a W-shaped distribution with a peak of 23.21 mm. For R_j = 0.25, the W-shape becomes more pronounced, and the peak settlement increases to 29.38 mm.

4.4. Genuine Pressure Analysis

The stress distribution in the tunnel lining under the influence of thinning defects is illustrated in Figure 18, based on engineering conditions with a burial depth of 34 m and Grade IV rock. The defect parameters are identical to those presented in Figure 14. On the outer side of the lining, the contact pressure between the lining and rock within the active thinning zone is significantly reduced, with localized contact failure observed in some areas. This indicates that the thinning defect weakens the interaction between the lining structure and the rock. Additionally, increased contact pressure is observed at the edges of the thinning zone, suggesting a stress redistribution wherein the edge regions bear a larger portion of the load. On the inner side, the primary lining completely loses contact with the secondary lining within the active thinning zone. In the thinning-affected zone, contact is maintained but becomes markedly non-uniform. The peak contact pressure is observed at the edge of the thinning zone on the inner lining surface, reaching 1.517 MPa, 1.452 MPa, 1.911 MPa, and 1.713 MPa under the defect scenarios shown in Figure 18a–d, respectively. Areas of poor contact may become potential sites for lining cracking or water ingress. The thinning defect significantly reduces the contact stiffness and force-transfer efficiency between the lining and the rock or secondary lining. This weakens the rock’s supporting function and exacerbates stress and deformation concentration in the lining, posing substantial risks to the long-term structural integrity and operational safety of the tunnel.

The contact pressure contour maps were used to extract the contact pressure curves along the axial direction of the thinning defect, which represent the genuine pressure profiles. The influence of different thinning parameters (P_j, L_j, L^arc_j, and R_j) on the genuine pressure curves is shown in Figure 19. To quantify the variation in genuine pressure caused by thinning, a dimensionless parameter—thinning-induced genuine pressure variation ratio (k_q)—is defined as the ratio of the difference between the genuine pressure in the thinning-affected zone $q_{\mathrm{thinning}}$ and that in the thinning-free region $q_{\mathrm{normal}}$ to the genuine pressure in the thinning-free region $q_{\mathrm{normal}}$, which is expressed as follows:

$$k_q={\displaystyle \frac{q_{\mathrm{thinning}}-q_{\mathrm{normal}}}{q_{\mathrm{normal}}}}\times 100\%$$

(1)

As shown in Figure 19a, the genuine pressure distribution induced by thinning exhibits a trough-like pattern, with the minimum pressure located at the center of the active thinning zone. Interestingly, the maximum genuine pressure does not occur at the edge of the thinning zone but rather within the thinning-affected zone. When P_j = 1, 3, 6, 8, or 9, the thinning-induced genuine pressure variation ratio (k_q) ranges from 2.43% to 7.72%. When P_j = 2, 4, 5, or 7, the k_q values are 10.23%, 17.01%, 14.96%, and 16.48%, respectively. Figure 19b shows that the thinning defect causes a reduction in genuine pressure within the active thinning zone and an increase within the thinning-affected zone. The maximum genuine pressure is consistently located in the thinning-affected zone. When L_j = 4 m, the maximum genuine pressure is 457.04 kPa. As L_j increases to 16 m, the maximum pressure increases to 525.83 kPa. For L_j = 20 m and 24 m, the maximum genuine pressures are 489.04 kPa and 562.19 kPa, respectively. According to Figure 19c, variation in the arc length L^arc_j significantly affects both the minimum and maximum genuine pressures. When L^arc_j = 4 m, the maximum and minimum pressures are 357.29 kPa and 41.31 kPa, respectively. As L^arc_j increases, the minimum pressure gradually increases, reaching 68.30 kPa when L^arc_j = 40 m. The corresponding maximum pressures for L^arc_j = 8, 16, 32, and 40 m are 478.80 kPa, 497.30 kPa, 441.14 kPa, and 425.39 kPa, respectively. As shown in Figure 19d, thinning defects significantly alter both the maximum genuine pressure in the thinning-affected zone and the minimum pressure in the active thinning zone. When R_j = 0.75, the maximum and minimum pressures are 122.36 kPa and 21.11 kPa, respectively. For R_j = 0.5, the corresponding values are 112.67 kPa and 11.05 kPa. When R_j is reduced further to 0, the maximum genuine pressure increases sharply to 482.36 kPa, indicating intensified stress transfer to adjacent areas.

In summary, thinning defects reduce the support capacity of the lining within the active thinning zone, shifting the load-bearing responsibility to the adjacent lining. As a result, genuine pressure decreases within the active thinning zone but increases significantly within the thinning-affected zone. Moreover, the thinning-induced degradation of the primary support may lead to a shift in the pressure mechanism from genuine pressure to loosening pressure. Consequently, the secondary lining—originally intended as a safety reserve—becomes a primary load-bearing component, which poses a serious threat to the long-term safety of tunnel operation.

5. Influence of Rock Lining Degradation on Rock Pressure with Thinning Defects

5.1. Degradation Characterization

With increasing service time, both the tunnel lining and rock in shallow HSR railway tunnels are prone to deterioration. To characterize the time-dependent degradation process of the lining, a degradation parameter is introduced, as proposed in [22], and expressed as follows:

$$E_{\mathrm{l}}(t)=E_{\mathrm{l}}^0[1-D_{e}t]e^{-\beta t}$$

(2)

$$E_{\mathrm{r}}\left(\mathrm{t}\right)=E_{\mathrm{r}}^0[1-\alpha \lg (t+1)]$$

(3)

where $E_{\mathrm{l}}(t)$ is the time-dependent elastic modulus of the lining; $E_{\mathrm{l}}^0$ is the initial elastic modulus of the lining; $D_e$ and $\beta$ are the damage and degradation coefficients of the support, respectively; $t$ is the service time of the tunnel; $E_{\mathrm{r}}\left(\mathrm{t}\right)$ is the time-dependent elastic modulus of the rock; $E_{\mathrm{r}}^0$ is the initial elastic modulus of the rock; and $\alpha$ is the degradation coefficient of rock elastic modulus.

5.2. Genuine Pressure Analysis Under Degradation Conditions

The genuine pressure distribution in shallow four-track HSR tunnels with thinning defects under conditions of lining or rock mass degradation is shown in Figure 20. The operational service times considered include 30, 60, 90, 120, and 150 years. When the tunnel lining deteriorates, the maximum genuine pressure increases with service time. At t_c = 0, the maximum genuine pressure is 112.67 kPa; when t_c = 150 years, it increases to 124.91 kPa—an increase of 10.86%. When only the rock deteriorates, a similar increasing trend is observed. At t_r = 150 years, the maximum genuine pressure reaches 246.28 kPa, representing an increase of 118.59 kPa compared to the initial state. When both the rock and the lining deteriorate simultaneously in the presence of a thinning defect, the genuine pressure rises further with time. At t_r = t_c = 150 years, the maximum genuine pressure reaches 262.16 kPa, reflecting a total increase of 132.69 kPa. In summary, as service time increases, the degradation of the lining and rock significantly exacerbates stress concentration within the thinning-affected zone, potentially resulting in severe structural damage. Moreover, such degradation may cause the tunnel to transition from a genuine pressure state to a loosening pressure state, with the secondary lining shifting from a safety reserve to a primary load-bearing structure. Therefore, in shallow four-track HSR tunnels, early identification and timely remediation of lining defects are essential to ensure long-term operational safety.

6. Conclusions

This study investigated the genuine pressure behavior of shallow four-track HSR tunnels affected by lining thinning and rock lining degradation through comprehensive field investigations, long-term monitoring, and numerical simulation. The main conclusions are as follows:

(1)

A field investigation conducted across ten HSR tunnels revealed a high prevalence of lining thinning defects, with 38.64% of the cases occurring at the vault. Thinning lengths were predominantly in the range of 4 to 8 m, accounting for 68.18% of the observed defects. The minimum lining thickness measured was as low as 0.09 m, and 84.11% of the defects had minimum thicknesses below 0.26 m. These findings indicate a substantial reduction in structural cross-section and highlight the potential risk to long-term tunnel performance and safety;

(2)

Field monitoring results from the XBS Tunnel indicate that the actual rock pressure is significantly lower than the values predicted by classical methods, including design code recommendations and Protodyakonov’s theory. Both settlement and convergence remained well below the design limits. Combined with the internal forces in the support, rock mass parameters, and the adopted support system, it was determined that the tunnel is primarily subjected to genuine pressure;

(3)

Lining thinning leads to significant stress redistribution and concentration, reduces the deformation resistance of the structure, and increases the likelihood of lining damage. It also alters the interaction between the lining and the surrounding rock or the secondary lining, resulting in non-uniform stress transfer and a notable increase in surrounding rock pressure. In shallow four-track HSR tunnels, the load-bearing mechanism may shift from genuine pressure to loosening pressure, and the secondary lining may transition from a safety reserve to a primary load-bearing component;

(4)

Time-dependent degradation of both the lining and surrounding rock further intensifies stress concentration. When both components degrade simultaneously, the genuine pressure can increase by more than 130 kPa. This degradation process accelerates the transition of the tunnel’s bearing mechanism—from genuine pressure to loosening pressure—in the presence of thinning defects.