Feasibility and Optimization Study on the Replacement of Core Rock Columns with Temporary Steel Supports in the Construction of Large-Section Subway Tunnels in Interbedded Rock Masses

Abstract

With the development of subway transportation, how to excavate large-section tunnels and find more convenient and reliable support methods has become an issue that cannot be ignored. This paper addresses issues such as low construction efficiency of core rock columns during the construction of large-section subway tunnels in sandstone–mudstone interbedded geological conditions. It proposes an optimized support scheme that replaces traditional core rock columns with temporary steel supports (steel columns). Finite element analysis was used to compare the deformation of the surrounding rock when retaining the core rock columns, using temporary steel columns to replace the core rock columns, and not providing additional support. Five interlayer positions and four interlayer angles were analyzed to identify the most dangerous geological conditions. Based on this analysis, the reasonable spacing of the temporary steel columns was investigated. The results indicate that temporary steel columns and core rock columns can effectively reduce vertical deformation of the surrounding rock, with steel columns showing slightly better results. Replacing core rock columns with steel columns is feasible. To control tunnel rock mass deformation, this project should ensure that the spacing between temporary steel columns is maintained between 21.88 m and 56.80 m. However, in construction sections with good rock mass conditions, the spacing can be extended as long as safety is ensured.

1. Introduction

In recent years, with the acceleration of overall urbanization, ground transportation has been unable to meet people’s daily travel needs, and this is particularly evident in relatively economically developed areas. As a high-capacity urban rail transit system, the subway can operate independently of ground transportation, providing people with punctual and efficient transportation services and effectively alleviating traffic pressure on the ground. Furthermore, with the advancement of the Western Development Strategy, the demand for and value of subway tunnels in western regions are gradually increasing. Compared with the eastern region, the western region has numerous mountains. Affected by geological processes such as plate movement and fault activity, the geological interlayering phenomenon is obvious in the western region. The topographical map of southwestern China is shown in Figure 1. Interbedded rock masses are characterized by heterogeneous rock properties and weak bedding structures [1,2], which result in complex stress conditions on the rock mass-support system during tunnel construction. This is particularly problematic for large-section shallow-buried tunnels, where support deformation, cracking, large plastic deformation, and even collapse are likely to occur [3,4]. Therefore, the construction safety of large-section tunnels under interlayered conditions cannot be ignored.

2. Aim of the Work

When constructing underground engineering projects in interbedded rock formations, the deformation and strength anisotropy characteristics of the rock layers, as well as the low strength characteristics of the structural planes, must be taken into consideration. These characteristics lead to poor stability of the tunnel rock mass, uneven stress distribution in the support structure, and even severe over-excavation or under-excavation during excavation. They also cause asymmetric deformation of the rock mass and severe tunnel bias [5,6,7,8,9]. Many scholars have conducted relevant research on these issues. Avci, M. et al. [10] investigated the causes of interbedding between limestone and mudstone in the eastern mountains of Turkey, while Boulanger, R.W. et al. [11] reevaluated the model parameters of the conventional liquefaction vulnerability index (LVI) method and nonlinear dynamic analysis (NDA) method, taking into account the impact of interbedding on the strength of composite strata. Przecherski, P. et al. [12] proposed a technical method for evaluating the failure conditions of interbedded sedimentary rocks. Through mesoscale numerical simulation, they studied the strength characteristics of interbedded samples with different bedding directions under the macro-anisotropic effects of interbedded sedimentary rocks. Eang, KE et al. [13] investigated the evolution and mixing mechanisms of limestone layered rock bodies through hydrogeochemical modeling and statistical analysis, while also studying the influence of groundwater on interlayer formation. Hua, L. [14] et al. described the rheological properties of soil using a fractional derivative-based Merchant model and simulated the drainage boundary of the pile–soil interface with a constrained drainage boundary. Wang, K. et al. [15] summarized theoretical research on the soil arching effect in pile-supported embankments, load distribution, soil deformation, as well as the structural factors influencing soil arching. Regarding the issues of surrounding rock deformation monitoring and prediction, Hu, D. et al. [16] proposed a ground settlement prediction method for rectangular tunnel construction based on complex functions and the Maxwell-Betti reciprocal work theorem, which demonstrated high reliability in practical engineering applications. Wang, M. et al. [17] proposed an InSAR-IPIM-based method for predicting three-dimensional surface deformation in backfill mining, which effectively corrects errors caused by using empirical parameters to predict NC1202 deformation. Li, Y. et al. [18] proposed an LSTM-ISA model by integrating time decay and a multiscale improved self-attention mechanism (ISA), which can serve as an effective tool for early warning of long-term pipeline settlement induced by rectangular shield tunnel construction.

Meanwhile, scholars worldwide have proposed numerous construction methods for the excavation of large-cross-section tunnels. In 1984, Japanese engineers employed the CD method during the construction of the Shinmei Highway Tunnel [19]. However, as tunnels constructed worldwide increasingly adopt flatter and larger-span designs, the technical requirements for construction methods have become progressively more demanding [20,21]. Iranian engineers explored the use of a central beam-column system for large-section cut-and-cover tunnel structures during the construction of the Tehran Metro in Iran and found that it worked well [22]. Italian scholars, on the other hand, focused on theoretical research, using finite element analysis to further study the interaction between tunnel initial support and surrounding rock, providing a reference for subsequent scholars [23]. Miura [24] innovatively proposed a method of first excavating a pilot tunnel and then expanding it for the New Tomei–Meishin large-section tunnel, which was applied to the project. This method is suitable for tunnels with good rock conditions and long routes. Zhang, J.Q. [25] points out that due to the large span and flat shape of the support structure, the bearing capacity it can provide is relatively small, which determines that large-section tunnel construction can only adopt sectional excavation. Therefore, the main construction methods suitable for large-section tunnels include the CD method, CRD method, single-side wall pilot tunnel method, and double-side wall pilot tunnel method. Many scholars have also conducted relevant research on tunnel steel structures. Yao Y. et al. [26] proposed the Steel-PEC Splice Frame Beam (SPSFB) in this paper, aiming to retain the advantages of PEC beams while reducing steel consumption. Shortly thereafter, Yao, Y. et al. [27] proposed a novel precast composite frame beam called the Reduced-Section Steel-Composite Frame Beam (RSCFB), which not only meets seismic resistance requirements but also reduces steel consumption without significantly increasing self-weight. Huang, H. et al. [28] designed and conducted push-out tests on partially encased composite columns (PECC) with shear studs arranged in different configurations, evaluating the effects of stud height, stud spacing, stud diameter, stud position, and concrete strength on the shear performance of PECCs.

Previous studies on subway station construction have extensively investigated large-section tunnel excavation and interlayer effects, covering excavation methodologies, interlayer spatial parameters (location/dip angle), and associated deformation mechanisms. These studies typically employ comparative case analyses to explore the correlations between construction techniques/interlayer conditions and the mechanical behavior of tunnels and surrounding rock. However, there has been little research on alternative solutions for core rock columns during excavation. When the reserved core rock columns experience a decline in bearing capacity due to long-term contact with groundwater and require reinforcement, there is insufficient research on solutions to problems such as reliance on external support, complex operations, and delays in construction progress. Therefore, finding a structural support component that can replace the core rock column is of great significance for the construction of large-section tunnels, temporary steel columns were erected on site as shown in Figure 2.

3. Project Overview

This study is based on the Chongqing Rail Transit Line 18 North Extension Project, with Kaixuan Road Station being the second station on the entire line. The design starting point of the Kaixuan Road Station is DK0+842.829, and the ending point is DK1+073.029, with a total length of 230.2 m, a cross-section excavation width of 23.42 m, an excavation height of 20.06 m, and an excavation area of 401.25 m². The main tunnel of the station involves multiple cross-section types. The station is a two-storey underground arched tunneled island station, and the tunnel is a straight-walled circular arch tunneled station cross-section, using a composite lining structure. The geological cross-section of the station is shown in Figure 3. The top layer is covered with mixed fill soil with a thickness of approximately 1–5 m. Below this, there are alternating layers of sandstone (blue area) and sandy mudstone (green area) with an interlayer dip angle of 8° to 16°. The rock type in the field area is interbedded sandstone and mudstone. The interfaces between sandstone and mudstone often exhibit mudification, especially in the upper sandstone and lower mudstone, where the interfaces are poorly bonded and classified as weak structural planes. The Chinese Railway Tunnel Design Code (TB 10003-2016) classifies rock mass grades [29]. When the Basic Quality Index (BQ) of the rock mass falls between 350 and 251, it is classified as Grade IV rock mass. When the BQ is less than or equal to 250, the rock mass is classified as Grade V rock mass. The tunnel’s surrounding rock classification is Grade IV. The working face primarily consists of sandstone with localized interbeds of sandy mudstone. The overburden thickness at the tunnel crown ranges from approximately 20 m to 29.6 m, with a burial depth of approximately 30.287 m to 33.059 m, classifying it as a shallow-buried tunnel.

The main tunnel excavation and support construction at Kaixuan Road Station commenced on 19 September 2023 and was completed on 25 April 2024. The on-site construction status is shown in Figure 4. In the 2024 Municipal Transportation Investment Party Committee Inspection Feedback Report, it was pointed out that the construction progress of the Chongqing Rail Transit Line 18 North Extension Project was unsatisfactory. Based on the current on-site construction situation, the excavation of the station’s main earthworks is expected to be completed by 30 March 2025, making it difficult to achieve the overall project schedule milestones. Following discussions among all parties involved in the project, temporary supports were added to the invert backfill surface between DK0+887.779 and DK0+893.779, and the core rock columns between DK0+917.729 and DK0+921.729 were removed. The plan is to complete the excavation of the main earthworks for the station by 20 January 2025, saving 70 days of construction time. This will also ensure the smooth passage of the shield machine through the section between Shibaotai Station and Kaixuanlu Station, meeting the project milestone requirements. This paper conducts research based on the above optimisation measures.

4. Research Plan and Feasibility Analysis

4.1. Model Construction

During the excavation of large-section tunnels, reserving core rock columns can form natural supports that resist the active earth pressure of the soil ahead, preventing large-scale collapse or deformation of the tunnel. Therefore, this is an extremely important step in the excavation of tunnels with poor geological conditions or large cross-sections. In actual engineering projects, it has been found that advancing the demolition time of core rock columns can accelerate project progress and reduce project costs. However, the early release of the core rock column will have a certain impact on the deformation of the tunnel surrounding rock. Therefore, if temporary steel columns can be used as temporary supports to replace the core rock columns, it will be possible to accelerate construction progress and reduce project costs. In addition, the materials used to construct temporary steel supports can be recycled after use, making full use of engineering materials and conserving natural resources.

When constructing the numerical model, a two-dimensional plane model was established based on the typical cross-section of DK0+887.779, as shown in Figure 5. In tunnel support systems, the initial lining is the most important support structure, so the role of secondary lining was not considered in the model simulation process.

The model was analysed using finite element software. The model has a horizontal dimension of 146 m, a vertical dimension of 130 m, and a tunnel depth of 32 m. The soil layers are divided into five layers, namely backfill soil, sandstone, sandy mudstone, sandstone, and sandy mudstone. The angle between the fourth and fifth soil layers is 11°, the contact type between interbedded layers in the model is the system-default Tied Contact. The distance between the left and right boundaries of the model and the tunnel is approximately 60 m, and is subject to horizontal constraints. The distance between the lower boundary of the model and the tunnel is approximately 70 m, and is subject to both horizontal and vertical constraints. The surrounding rock was modelled using the Mohr–Coulomb constitutive model, while the initial support was modelled using an elastic constitutive model and simulated using 2D plane elements.

The model parameters were determined by comprehensively considering the on-site geological survey report, the reference values for rock mass and structural parameters specified in the ‘Code for Geotechnical Investigation of Urban Rail Transit’ (GB 50307-2012) [30], and relevant references [31,32]. The specific model parameters are listed in

Table 1. Mechanical properties of materials.

Category	Severe (KN/m3)	Elastic Modulus (MPa)	Poisson’s Ratio	Friction Angle	Cohesion (kPa)	Material Type	Property
Backfilling Soil	18.0	5	0.33	15	20	Mohr–Coulomb	Physical Unit
Limestone	24.9	3564	0.14	40.7	1560	Mohr–Coulomb	Physical Unit
Sandy Mudstone	25.6	1431	0.37	32.6	640	Mohr–Coulomb	Physical Unit
Steel Support	78.5	210,000	0.3	/	/	Elastic Unit	Plate Unit (Computing)
Initial Support	25	10,000	0.2	/	/	Elastic Unit	Plate Unit (Computing)

.

4.2. Simulation Plan

For the excavation of large-section tunnels, the main structure uses a double-sided wall pilot tunnel method, with three steps and nine pilot tunnels excavated. As shown in Figure 6, for ease of analysis, the nine access tunnels were numbered. The excavation of the station structure was divided into three stages. First, the upper and middle stages of the station were constructed, followed by the excavation of the lower stage. The two side access tunnels were excavated first, followed by the middle access tunnel. The detailed simulation process is shown in Figure 7.

The tunnel excavation process is illustrated in Figure 7. To reduce computational requirements during simulation, some construction steps have been merged or simplified:

1. Excavate the upper-right pilot tunnel ②, install primary support, and set up intermediate temporary support.

2. Excavate the upper-left pilot tunnel ① and the right-middle pilot tunnel ④, establish primary support, and install intermediate temporary support. Meanwhile, remove the support at the bottom of pilot tunnel ② to facilitate excavation.

3. Excavate pilot tunnels ③ and ⑥, establish primary support, and install intermediate temporary support. Meanwhile, remove the support at the bottom of pilot tunnels ① and ④ to facilitate excavation.

4. Excavate pilot tunnels ⑤ and ⑦, establish primary support, and install intermediate temporary support. Meanwhile, remove the support at the bottom of pilot tunnel ③ to facilitate excavation.

5. Excavate pilot tunnel ⑧ and establish primary support. Meanwhile, remove the support at the bottom of pilot tunnel ⑦ to facilitate excavation.

6. Excavate pilot tunnel ⑨ and establish primary support. Meanwhile, remove the support at the bottom of pilot tunnel ⑧ to facilitate excavation.

The tunnel excavation cycle advance is 2.5 m. After the cross-section deformation stabilizes, the double-sidewall support is dismantled sequentially. During excavation, the location of the reserved core rock column is determined based on site requirements, or temporary steel columns are installed at the reserved core soil location as an alternative. Here, no additional support is defined as Condition 1, reserved core rock column as Condition 2, and temporary steel columns as Condition 3.

5. Tunnel Deformation Analysis

5.1. Maximum Rock Mass Deformation Analysis

The main role of the core rock column in the overall tunnel structure is to control tunnel deformation, particularly vertical deformation, such as controlling crown settlement and arch bottom uplift. As shown in Figure 8, the vertical deformation of the surrounding rock under the three operating conditions is mainly concentrated at the arch crown and arch base, manifesting as subsidence and uplift. Furthermore, overall observation shows that the deformation of the rock mass around the tunnel is asymmetrical, which is more evident in the a cloud diagram without additional support and the b cloud diagram with the remaining core rock column. This is due to the angular asymmetry of the strata [33]. Under three operating conditions, the settlement of the tunnel crown is generally around 7 mm, but the uplift of the tunnel floor exceeds 19 mm in all cases. This is determined by the rock strata in which the tunnel is located. Since the interbedded strata are positioned at the arch waist of the tunnel, the section above the arch waist consists of sandstone, which has good overall rock properties and is less prone to deformation. However, the arch bottom section is situated within sandy mudstone strata, classified as Grade V rock mass, which has poorer stability and thus experiences greater deformation. Overall observation of the vertical displacement cloud map reveals that rock deformation is relatively minor when temporary steel columns are erected and core rock and soil are reserved, but the difference is not significant compared to situations where no additional support is provided.

In numerical simulations, it was observed that in all three cases, the maximum deformation of the tunnel occurred at the crown and base of the tunnel after construction was completed. The maximum settlement occurred at the crown, marked as point A, and the maximum uplift occurred at the base, marked as point B, as shown in Figure 9.

The vertical deformation diagrams of the surrounding rock for the three excavation schemes are shown in Figure 10. As can be seen from the figure, both the arch crown settlement and the arch base uplift deformation can be divided into two development stages: rapid deformation, in which the deformation rate increases rapidly in the first 20 steps; and slow deformation, in which the deformation rate increases slowly from the 20th to the 47th step. In the figure, we have used yellow lines to distinguish between the two developmental stages and marked the maximum deformation values for each stage with green lines. During the ‘rapid deformation’ stage, the primary support and surrounding rock jointly bear the stress redistribution caused by rock excavation, resulting in joint deformation. The significant deformation during this stage is due to the use of flexible primary support, which maximises the self-supporting capacity of the surrounding rock and reduces the pressure on the primary support. ‘Slow deformation’ refers to the disturbance deformation caused by subsequent excavation. In the three excavation schemes, the tunnel crown settlement and arch bottom uplift reached 6.09 mm and 18.85 mm, respectively, during the ‘rapid deformation’ stage, indicating that the maximum value of the tunnel’s final vertical deformation mainly originates from the accumulation during the ‘rapid deformation’ stage.

However, as shown in Figure 10a, the final crown settlement values of the three excavation schemes still differ. Retaining the core rock columns and installing temporary steel columns can reduce the maximum crown settlement by 0.07 mm and 0.27 mm, respectively, while the maximum arch bottom uplift is reduced by 0.40 mm and 0.36 mm, respectively. However, the vertical deformation values of the three excavation schemes differ only slightly. This is because, in the construction simulation process, the reserved core rock columns and the temporary steel columns are both located in the middle of the tunnel, approximately 50 m from the simulated tunnel starting point. In the simulation, due to the distance limitations of the core rock columns and temporary steel columns in supporting the tunnel rock mass, their effect on the rock mass decreases as the distance from them increases. Compared to the excavation scheme without additional support, although the two excavation schemes can control the maximum vertical deformation, their control effectiveness is poor, and the reduction in deformation is small.

5.2. Analysis of Rock Mass Deformation Around Structural Support Components

Here, the surrounding rock at distances of 5 m, 10 m, and 15 m from the reserved core rock columns and the temporary steel columns was selected for analysis. As shown in Figure 11a–c, as the distance increases from 5 m to 15 m, the maximum arch crown settlement reduction of Excavation Scheme 2 compared to Excavation Scheme 1 decreases from 9.22% to 4.13%, and the maximum arch crown settlement reduction of Excavation Scheme 3 compared to Excavation Scheme 1 decreases from 31.49% to 13.68%. Additionally, as the distance increases, the maximum arch crown settlement values under the three excavation schemes gradually converge. As can be seen from Figure 11d–f, the further away from the structural support components, the greater the arch bottom uplift values for Excavation Scheme 2 and Excavation Scheme 3, which are approximately close to the maximum arch bottom uplift value of Excavation Scheme 1. Furthermore, the magnitude of deformation is greatest for Excavation Scheme 1, followed by Excavation Scheme 2, while Excavation Scheme 3 consistently exhibits the smallest deformation values.

Based on the above data results, it can be concluded that reserving core rock columns and erecting temporary steel columns play a crucial role in controlling vertical deformation of the tunnel surrounding rock and ensuring its stability. Furthermore, the control effect of erecting temporary steel columns on the surrounding rock is stronger than that of reserving core rock columns, indicating that erecting temporary steel columns can be used as an alternative to core rock columns. This approach also helps reduce interference between the construction of core rock columns, middle and lower benches, and the construction of the invert and secondary lining. However, it is important to note that when using temporary steel columns for support, the effectiveness of the temporary steel columns will gradually decrease as the distance from the structural support components increases. Therefore, during the engineering application process, it is necessary to determine the number of temporary steel columns or the distance between the temporary steel columns and the working face or secondary lining to prevent excessive tunnel deformation and the resulting risks such as collapse.

5.3. Analysis of the Rationality of Temporary Steel Columns

In the design and actual construction plan, the entire temporary steel column consists of six main steel columns, as well as horizontal tie beams and diagonal braces used for connection. As can be seen from the stress distribution of the temporary steel columns in Figure 12a,b, only the horizontal tie beams and diagonal braces used to connect the steel columns are subjected to tensile stress, while the steel columns as a whole are mainly subjected to compressive stress. It is also evident that steel columns closer to the working face are subjected to greater compressive stress. This is primarily because as excavation continues, the original stress state of the rock mass is disrupted, causing the rock mass behind the working face to begin moving toward the tunnel cavity, resulting in surrounding rock deformation. The steel supports closer to the working face are the first and most strongly resist this deformation, bearing the greatest initial load.

During tunnel excavation, the maximum tensile stress experienced by the temporary steel columns was 14.85 MPa, while the maximum compressive stress was 65.74 MPa. The temporary steel columns were constructed using 609 steel, which has a tensile strength exceeding 490 MPa and a yield strength exceeding 355 MPa. This indicates that the tensile and compressive stresses experienced by the temporary steel columns are far below the tensile strength and yield strength of the steel, meaning that the steel itself will not undergo irreversible damage. Furthermore, in projects where the rock strata strength is relatively high, the strength of temporary steel columns can be further reduced based on the actual project conditions (e.g., by using steel types with lower strength or reducing the size of the support structures), thereby further reducing project costs while ensuring safety.

5.4. Numerical Model Validation

To validate the feasibility of the numerical model, this paper relies on an actual Chongqing Metro tunnel project and relevant design data to select a tunnel cross-section with temporary steel columns for arch crown settlement monitoring (the on-site monitoring system is shown in Figure 13) and compares and analyses the results with the numerical simulation results. The selected location is at the front end of the tunnel, similar to condition 3. Data from a measuring point 10 m away from the temporary steel column was selected for comparison. The measuring point number is DK0+871-A3B. The monitoring frequency is adjusted in a timely manner according to different construction periods and monitoring results, while eliminating random errors in the data.

The monitoring period lasted a total of 71 days. The on-site monitoring results were compared with the simulation results of operating condition 3 for verification, as shown in Figure 14. From the graph, the trend of the line chart of the numerical simulation values and the actual monitoring values shows a certain degree of similarity, but in terms of the magnitude of change, the actual monitoring results are slightly greater than the numerical simulation results. The reason may be that the actual construction process spans a long period of time, and factors such as interlayer cracks and soft rock creep can affect the deformation of the surrounding rock. From the results, the maximum arch settlement in the numerical simulation was 5.65 mm, while the maximum arch settlement in the actual monitoring was 6.20 mm. The maximum difference between the numerical simulation and the monitoring results was only 0.55 mm, which is not significant. This indicates that the finite element model and calculation results of the numerical simulation are reasonable.

6. Analysis of the Optimal Spacing for Temporary Steel Columns

During tunnel excavation, the deformation of the tunnel surrounding rock is influenced by numerous factors. Among these, the location of interlayers and differences in interlayer angles can significantly impact the stability of the tunnel surrounding rock. Therefore, before determining the optimal spacing for temporary steel columns, it is essential to identify the most hazardous geological conditions of the project. Conducting research based on this information will better ensure construction safety and provide valuable reference for on-site construction activities.

6.1. Analysis of the Impact of Interlayer Position on Surrounding Rock Deformation

This tunnel section is 230.2 m long, as shown in Figure 3. The rock layers at the tunnel face exhibit alternating layers of sandstone and mudstone. Therefore, this tunnel section is divided into five categories based on the relative positions of the alternating layers with respect to the tunnel cross-section. These categories are: alternating layers above the tunnel cross-section, interbedded layers crossing the crown of the tunnel cross-section, interbedded layers crossing the waist of the tunnel cross-section, interbedded layers crossing the base of the tunnel cross-section, and interbedded layers located below the tunnel cross-section. The relative positions are shown in Figure 15 below. To control variables, the upper three soil layers were not modified, maintaining the maximum thicknesses of the fill soil, sandstone, and mudstone at 4.1 m, 20.5 m, and 10.0 m, respectively, with the interlayer angle kept at 11°, consistent with the actual soil layer conditions in the middle section of the tunnel. However, the ratio of sandstone to mudstone differs from interlayer position 1 to interlayer position 5, the ratio of sandstone to mudstone is 0 (entirely within the mudstone layer), 0.3 (interlayers crossing the arch crown), 0.6 (interlayers crossing the arch waist), 0.9 (interlayers crossing the arch base), and 1 (entirely within the sandstone layer).

Tunnel excavation was also carried out using the double-sided wall pilot tunnel method. After the five interlayered positions were excavated, the maximum arch crown settlement and maximum arch base uplift are shown in Figure 16. As can be clearly seen from Figure 16a, as the position of the sandstone–mudstone interlayer gradually moves downward, the proportion of sandstone layers in the tunnel cross-section gradually increases, and the maximum settlement of the tunnel crown gradually decreases. Compared with interlayer position 1, the maximum settlement of interlayer position 5 decreased by 30.93%. The settlement value at Interlayer Position 1 in the first step is significantly greater than that of other conditions because during the construction phase, the first step of excavation was carried out at the upper right guide tunnel 2. The tunnels at Interlayer Position 1 are all located in mudstone strata with poor rock quality, while the upper right guide tunnel 2 at the other interlayer positions are all located in sandstone strata. Therefore, the settlement value of the excavation at Interlayer Position 1 differs significantly from that of other interlayer positions in the initial stage. Under the condition of considering only the arch settlement, it can be determined that the stability of the surrounding rock at the five interlayer positions decreases gradually from interlayer position 5 to interlayer position 1.

As shown in Figure 16b, the arch bottom uplift values of interlayer positions 1, 2, and 3 are similar, at 19.65 mm, 19.56 mm, and 19.63 mm, respectively. This indicates that when interlayers are located at the arch waist or above, the interlayer position has little effect on the arch bottom uplift, while the arch bottom uplift is mainly influenced by the properties of the surrounding rock near the arch bottom. However, sandstone is present in the rock surrounding the arch bottom at interlayer positions 4 and 5, and the rock mass properties are good, so the arch bottom uplift value is smaller. Therefore, under the condition of considering only the arch bottom uplift, it can be determined that the stability of the surrounding rock in the five working conditions is ranked as follows: interlayer position 5, interlayer position 4, interlayer position 3, interlayer position 2, and interlayer position 1, with the stability of the surrounding rock in interlayer positions 3, 2, and 1 being similar. Taking into account both the arch crown settlement and the arch base uplift, it can be concluded that the rock mass stability at Interlayer Position 1 is the poorest. The maximum arch base uplift and the maximum arch crown settlement are both the largest among the five operating conditions.

6.2. Analysis of the Effect of Interlayer Inclination on Rock Mass Deformation When the Interlayer Is Located Above the Tunnel Cross-Section

The dip angle of interbedded rock bodies is one of the key geological factors controlling the stability and deformation patterns of tunnel surrounding rock. Different inclination angles can significantly affect the strength, anisotropy, failure mechanisms, and load characteristics of rock masses, which in turn lead to different deformation characteristics in tunnels. The dip angle of the sandstone–mudstone interbeds in this tunnel ranges from 8° to 16°. Therefore, to investigate the effect of interbed dip angle on tunnel deformation, based on the research in Section 4.1, the state of the top three layers of rock and soil remains unchanged. At this point, the interbedded position is above the tunnel cross-section. Additionally, the dip angle between the fourth and fifth layers of sandstone and mudstone is adjusted into four groups: 8.5°, 11°, 13.5°, and 16°. The tunnel deformation results obtained from numerical simulation are shown in Figure 17.

As can be seen from the figure, the four interlayered inclinations have little effect on the arch bottom uplift, with the maximum arch bottom uplift being around 19.65 mm and the difference being less than 0.01 mm. Therefore, this study focuses on analysing the impact of interlayer inclination on the maximum settlement of the tunnel crown. As shown in Figure 17, as the interlayer inclination increases, the maximum settlement of the tunnel crown also gradually increases. When the interlayer inclination is 16°, the maximum settlement of the tunnel crown exceeds that of an interlayer inclination of 8.5° by 14.77%. Furthermore, since the differences in the dip angles of the four interlayers are not significant, the differences in the maximum subsidence values of the roof are also small. When the dip angles of the interlayers increase further, the changes in subsidence will become more pronounced. In addition, only the angle of the interbedded zone was considered during the simulation. To find the most dangerous stratum conditions, the right end of the model was used as the rotation centre when changing the model inclination. As the interbedded inclination increased, the thickness of the rock strata also changed. When the interbedded inclination reached 16°, the thickness of the sandstone layer reached its minimum, and under the combined influence of the interbedded inclination and layer thickness, the tunnel deformation reached its maximum.

6.3. Analysis of the Optimal Spacing of Temporary Steel Columns

6.3.1. Spacing Analysis Under the Worst-Case Scenario

The geological conditions of a tunnel are influenced by various factors, including mudstone content, interbedding dip angle, and interbedding thickness. As indicated by the simulation results and relevant studies [34], the higher the mudstone content, the poorer the overall performance of the interbedded rock mass; when the interbedding dip angle is less than 45°, the larger the dip angle, the poorer the overall performance of the interbedded rock mass; the greater the interlayer thickness, the more the interlayer rock mass is influenced by the properties of the mudstone, resulting in poorer overall performance. Therefore, considering all factors, an interlayer angle of 16° and an interlayer location above the arch crown were selected as the study conditions. Under these conditions, the tunnel cross-section is entirely within mudstone strata, resulting in poorer overall stability.

Based on the spacing between the temporary steel columns, simulation schemes 1, 2, 3, 4, 5, and 6 are defined, with the spacing between the temporary steel columns being 15 m, 17.5 m, 20 m, 22.5 m, 25 m, and 27.5 m, respectively. The simulation scheme construction diagram is shown in

Table 2. Simulated construction diagram showing different spacing.

Simulation Plan	Spacing	Schematic Diagram
1	15 m
2	17.5 m
3	20 m
4	22.5 m
5	25 m
6	27.5 m

. The tunnel is 97.5 m long. The temporary steel columns consist of six columns, several horizontal tie beams and cross braces, measuring 7.2 m in length and 4.2 m in width. The tunnel is not completely connected but leaves 2.5 m of rock as the working face to simulate the actual construction site.

As shown in Figure 18a,b, as the spacing between steel columns increases, the control force on the surrounding rock decreases, and both the maximum arch crown settlement and the maximum arch base uplift exhibit an upward trend. Compared to Simulation Scheme 1, Simulation Scheme 6 shows an increase of 21.00% in the maximum arch crown settlement and 12.90% in the maximum arch base uplift. In accordance with on-site construction requirements and the ‘Technical Specifications for Monitoring of Urban Rail Transit Engineering’ (GB50911-2013) [35] and the ‘Interim Measures for Third-Party Monitoring Management of Chongqing Rail Transit,’ tunnel deformation warnings are classified into three hazard levels based on severity: Level I, Level II, and Level III. Detailed information is provided in

Table 3. Hazard Classification Table.

Alert Level	Warning Conditions	Early Warning Measures
Ⅰ	Cumulative Deformation Value < 17 mm	Real-time Monitoring
Ⅱ	17 mm ≤ Cumulative Deformation Value < 20 mm	Send Early Warning Messages; Conduct Encrypted Monitoring; Take Measures such as Optimising Support.
Ⅲ	Cumulative Deformation Value ≥ 20 mm	Activate the Emergency Response Plan and Suspend Construction.

.

When deformation is less than 17 mm, it falls under Grade I hazard level, at which point only monitoring is required, and no additional support measures are needed. Therefore, for all six simulated scenarios, the maximum arch crown settlement does not exceed 9 mm. Since the arch crown overlying rock is sandstone, which is less prone to deformation compared to mudstone, under the condition of considering only arch crown settlement, the smaller the spacing between temporary steel columns, the smaller the deformation of the arch crown overlying rock. However, when the spacing is extended from 15 m to 27.5 m, the overlying rock conditions remain relatively safe.

The surrounding rock near the arch bottom is mudstone, with poor rock properties. As shown in Figure 18b, when the spacing between temporary steel columns is less than 20 m, it is classified as Grade I hazard level, while when the spacing is between 20 m and 27.5 m, the overall surrounding rock has reached Grade II hazard level. When the uplift value exceeds 17 mm, reinforcement measures must be taken to ensure construction safety. Therefore, the spacing between temporary steel columns at the construction site should be controlled between 17.5 m and 20 m to prevent excessive deformation at the arch bottom location.

Since this project involves a large-section shallow-buried tunnel located in an alternating layer of hard rock on top and soft rock below, the deformation of the tunnel arch bottom is much greater than that of the arch top. The selection of the optimal spacing also depends on the extent of the arch bottom uplift. Therefore, if the arch bottom is reinforced to further control deformation, the spacing between temporary steel columns can be appropriately increased while ensuring safety. At the same time, for projects with better rock conditions, the spacing between temporary steel columns can be selected based on actual conditions. Replacing core rock columns with temporary steel columns as early as possible not only saves construction time but also facilitates the advancement of secondary lining. Additionally, the materials used for temporary steel columns can be recycled, thereby avoiding waste of steel.

6.3.2. Spacing Analysis for All Stratigraphic Conditions

Here, the safety factor K is introduced based on the most unfavorable stratum conditions, and two related proportionality coefficients are defined as M and N.

$$\begin{array}{c}M=m_0\times m_1\end{array}$$

(1)

In this model, M serves as the dip angle safety factor, where $m_0$ quantifies the deformation control effectiveness for vault settlement across varying dip angles, while $m_1$ correspondingly evaluates the deformation control performance for invert heave under different angular conditions.

$$\begin{array}{c}N=n_0\times n_1\end{array}$$

(2)

In this analysis, N represents the sandstone–mudstone ratio safety factor, where $n_0$ quantifies the deformation control effectiveness for vault settlement under varying sandstone–mudstone ratios, and $n_1$ correspondingly evaluates the deformation control performance for invert heave under different lithological conditions. Based on these parameters, the overall safety coefficient K can be expressed as follows:

$$\begin{array}{c}K=M\times N\end{array}$$

(3)

According to the statistical results in Section 6.1 and Section 6.2, the range of values for the safety factor K at this construction site is obtained, as shown in Figure 19.

Taking the interlayer angle of 16° and the interlayer position above the arch as the reference condition, the safety factor is set to 1. As the interlayer position moves upward and the interlayer inclination angle decreases, the safety factor K gradually increases to a maximum of 9.14. At this point, the interlayer inclination angle is 8.5°, and the entire tunnel cross-section is located in sandstone strata, achieving the highest level of safety. Therefore, the spacing between steel columns during tunnel excavation is summarised as follows:

$$\begin{array}{c}L=K\times L_0\end{array}$$

(4)

where L is the optimal spacing of steel columns under the given geological conditions, K is the safety factor, and L_0 is the optimal spacing of steel columns under the worst-case scenario.

Here, the interlayer dip angle is taken as 11°, and the optimal spacing of steel columns is verified under the condition that the interlayer passes through the arch waist position. According to empirical formula 4, it can be inferred that the spacing should be maintained between 32.55 m and 37.20 m. When the spacing between steel columns is 32.5 m, the maximum arch bottom uplift in numerical simulation is 17.42 mm; When the spacing is 37.5 m, the maximum arch bottom uplift in the numerical simulation is 16.53 mm, consistent with the formula’s conclusion. Therefore, it is considered that this empirical formula is relatively reasonable. Therefore, the reasonable range for temporary steel columns in this project should be maintained between 17.50 m and 182.80 m. However, since the main body of the tunnel is located in an area of alternating sandstone and mudstone layers, and the cross-section is situated within sandstone layers with only a very small interval between mudstone layers, the K value is taken to be between 1.25 and 2.84 when excluding extreme cases. This means that the spacing between temporary steel columns should be maintained between 21.88 m and 56.80 m.

6.4. Application at Construction Sites

The original plan was to complete the excavation of the main earthworks for the station by 30 March 2025. However, due to unsatisfactory construction progress on site, it was difficult to meet the overall project schedule targets. Based on the above simulations and research by all parties, it was decided on site to add temporary supports and remove the original core rock and soil. In reality, the main earthworks excavation will be completed by 20 January 2025, saving a total of 70 days of construction time. This will also ensure that the shield machine can pass through the station smoothly, meeting the milestone target requirements. After considering the opinions of all parties, the construction site appropriately reinforced the tunnel support, especially the arch bottom section. The design ensured that the longitudinal distance between the temporary steel columns and the secondary lining ring was less than or equal to 45 m, which was consistent with the simulation results. To reduce interference between the construction of the middle and lower steps and the cross-construction of the invert and secondary lining, after discussion and research by all parties involved in the project, it was decided in December 2024 to adjust the longitudinal distance between the secondary lining of the ring and the temporary steel columns to less than or equal to 75 m.

7. Discussion

The deformation of rock mass surrounding large-section tunnels in interbedded rock bodies is influenced by the combined effects of numerous factors. Research into their deformation and support methods is a meaningful topic, but due to various limitations, there are still many shortcomings in the research process.

(1) During the finite element simulation, factors such as the properties of the surrounding rock strata, tunnel depth, and cross-section dimensions were not considered. Additionally, the contact types between interlayers may also affect the simulation results. Subsequent studies could conduct multi-factor numerical simulations to further investigate their coupling effects and determine the optimal reinforcement strategy.

(2) In this paper, temporary steel columns were erected immediately after excavation was completed. However, the time required to erect temporary steel columns affects on-site construction costs. Therefore, the timing of steel column erection can be discussed in terms of optimal on-site benefits to achieve the lowest construction costs.

(3) Simulations of interlayer dip angles and interlayer thicknesses are relatively limited. Based on this, we plan to conduct physical model tests on sandstone–mudstone interlayered rock bodies and use the test conclusions to further verify the simulation results. At the same time, we will combine actual field conditions to further refine the theoretical formula for calculating the optimal spacing of steel columns to achieve wider application.

8. Conclusions

The study was conducted using the Chongqing Metro Line 18 as a case study, employing finite element simulation methods to investigate the feasibility of using temporary steel columns as a substitute for core rock and soil. Based on this, the study further explored the optimal spacing of temporary steel columns under the most hazardous geological conditions. The results of the study indicate that:

(1) During the excavation of large-section tunnels, temporary steel columns can reduce crown deformation by more than 9.96% compared to core rock columns. Although the improvement in the control of arch bottom uplift is not significant, it is still within the safe range. In addition, the use of temporary steel columns instead of core rock columns can effectively shorten the construction period and reduce construction costs. Therefore, the use of temporary steel columns instead of core rock columns is an effective construction method.

(2) During the excavation of large-section tunnels in interbedded rock formations, the position and angle of the interbeds significantly affect the deformation of the surrounding rock. The higher the mudstone content in the tunnel cross-section and the greater the interbed angle, the poorer the overall stability of the surrounding rock and the greater the deformation. Based on the worst-case geological conditions of this project, to ensure construction safety, the spacing between temporary steel columns should be controlled between 17.5 m and 20 m. However, during actual construction, if measures such as increasing support strength or excavating tunnel sections with relatively better surrounding rock conditions are taken, the spacing between steel columns can also be extended to reduce engineering costs.

(3) For the entire project, this paper proposes a selection scheme for the spacing of steel columns. The optimal spacing between steel columns generally satisfies the formula L = K × L_0, where L represents the optimal spacing of steel columns under the given geological conditions, K is the safety factor, and L_0 is the optimal spacing of steel columns under the most unfavourable conditions. Based on this empirical formula and excluding extreme geological conditions, the value of K ranges from 1.25 to 2.84, and the spacing between steel columns should be maintained between 21.88 m and 56.80 m, which is consistent with the on-site construction plan.