Study on the impact of new pier cap construction on adjacent high-speed railway subgrades

2.1. Project overview

The new intercity railway under study is located in the Yangtze River Delta Plain, with a design speed of 350 km/h and a total length of 278.53 km. The route runs parallel to the existing Shanghai-Nantong HSR at multiple points, where some pier caps are constructed within 30 m of the existing line-qualifying as adjacent construction. A total of 305 pier cap structures are involved. This study focuses on a typical case where the minimum excavation distance is only 0.5 m from the existing subgrade retaining wall, and the maximum excavation depth reaches 5 m. The retaining wall is embedded 12 m deep, with its base at –4 m elevation. The HSR subgrade comprises a 4.5 m-high fill with a 2:3 slope on the left and a 0.5 m gravel cushion beneath, supported by 1 m-diameter, 70 m-deep bored piles spaced 2.5 m apart. The top supports six-track high-speed rails. A 0.5 m-thick retaining wall is located on the right. The excavation is supported by 12 m-long steel sheet piles with waling installed at 0.5 m depth. The pier foundation consists of 4×2 bored piles (1 m diameter, 68.5 m long), with the pile cap bottom at −1.5 m elevation and a height of 2.5 m. The cap base is sealed with 0.5 m-thick C15 concrete. The groundwater level is at 0 m. See Fig. 1 for the schematic layout.

2.2. Soil parameters

The existing high-speed railway line is constructed using embankment fill and a gravel cushion. During later construction stages, miscellaneous fill is used for backfilling the excavation pit. The soil at the location of the new pier caps primarily consists of silty clay and silty clay with silt. The basic parameters of each soil layer are listed in Table 1.

Construction activities at the pier cap locations involve excavation and backfilling, representing a process of soil unloading and reloading. These excavation processes induce small-strain behavior, under which the soil exhibits a relatively high elastic modulus. The Hardening Soil Small-Strain (HSS) model provides a more comprehensive representation of the nonlinear and stress-path-dependent deformation behavior of soils, making it particularly suitable for accurately predicting the environmental impact of excavation [14]. The parameters adopted for the HSS model are listed in Table 2.

Concrete structural elements are modeled using a linear elastic approach. The embankment fill, gravel cushion, and backfilled soils are simulated using the Mohr-Coulomb (M-C) model. Both the new structures and existing HSR pier caps and columns are constructed with C30 concrete and HRB400 reinforcement.

2.3. Structural component parameters

The cofferdam is constructed with Larsen IV-type steel sheet piles. Retaining walls and concrete bedding are modeled as shell elements, while the support structure uses beam elements (Q235 double H-section steel). Bored piles are simulated as embedded pile elements (C30 concrete, HRB400 reinforcement). Key material parameters are summarized as follows: the steel sheet pile has a thickness of 0.17 m and elastic modulus of 2.06×10⁸ kPa; the retaining wall and cushion layer have thicknesses of 0.5 m with elastic moduli of 3×10⁷ kPa and 1.476×10⁷ kPa respectively; the pile elements have a diameter of 1.0 m and elastic modulus of 3.0×10⁷ kPa.

2.4. Model development

Given the uniform layout of the existing embankment fill and pile foundations in the direction perpendicular to the Shanghai-Tong high-speed railway, along with their consistent cross-sectional geometry, stress conditions, and loading mechanisms, the system is simplified into a 2D plane strain problem in PLAXIS. The most critical scenario is selected for numerical simulation. For ease of analysis, a Cartesian coordinate system is adopted, with the borehole location set as the origin (0, 0). The model boundaries span from –20 m to 50 m along the $x$-axis and from –100 m to 20 m along the $y$-axis. To enhance calculation accuracy, 15-node triangular elements are used for soil meshing. The groundwater table is set at an elevation of 0 m, and flow is restricted at the model boundaries. The final finite element model consists of 2,835 elements and 25,501 nodes. The mesh layout of the finite element model is illustrated in Fig. 2. The model is validated against field monitoring data, with deviations within 15 %.

Fig. 1Site engineering dimension drawing (m)

Fig. 2Mesh division of the model

3.1. Subgrade soil displacement

Six monitoring points (A to F) were selected. Horizontal and vertical displacement curves were plotted, as shown in Fig. 3. Displacement diminishes with increasing distance from the excavation.

As shown in Fig. 3, dewatering and excavation during the early construction stages represent a typical unloading process, causing horizontal displacements toward the pit and settlements due to groundwater drawdown reducing pore pressure and increasing effective stress. The maximum horizontal displacement is 9.71 mm and maximum settlement is 9.73 mm, decreasing with distance from the pit. Activation of the support structure and construction of the concrete bedding layer induce slight reverse horizontal displacements in remote sections of the subgrade, mainly due to the lateral expansion force exerted by the transverse bracing system. The greatest displacement increment (3.94 mm) is caused by cofferdam removal, followed by support system and pile activation. All construction stages result in vertical settlement of the subgrade soil, with pier construction and loading most significantly influencing settlement.

Fig. 3Subgrade soil displacement curves

a) Horizontal displacement

b) Vertical displacement

3.2. Displacement of subgrade pile foundations

The maximum displacement of each pile occurs at its top node. Eleven piles were labeled sequentially from 1 to 11 based on proximity to the excavation pit. Horizontal and vertical displacement curves of the pile tops are shown in Fig. 4.

Fig. 4Displacement curves of pile foundations

a) Horizontal displacement

b) Vertical displacement

As shown in Fig. 4, horizontal displacement of the surrounding soil causes lateral deflection and settlement of the piles, decreasing with distance from the pit. Pile 1, closest to the excavation, shows maximum horizontal displacement of 6.08 mm and settlement of 4.56 mm. Sheet pile removal has the greatest impact on pile horizontal displacement, followed by drilling and support system operations. For vertical displacement, drilling and pier column casting have the most pronounced effects.

3.3. Influence of construction distance on subgrade displacement

The distance from the new pier cap to the existing subgrade is denoted as $d$. Keeping all other conditions constant, $d$ was varied as follows: 0.25 m, 0.5 m, 1 m, 2 m, 4 m, 6 m, 8 m, 10 m, 12 m, 14 m, 16 m, 18 m, 24 m, 32 m, 48 m, and 64 m. For each case, the maximum horizontal and vertical displacements of the existing subgrade soil were determined. The point of maximum displacement consistently occurred at the highest elevation of the embankment fill. Empirical relationships between displacement ($u_x$, $u_y$) and distance ($d$), as shown in Fig. 5. Both horizontal and vertical displacements clearly decrease with increasing $d$.

The fitting equations derived from the data are as follows:

(1) Horizontal displacement:

where $u_x$ is the horizontal displacement (mm), $d$ is the distance from the new pier cap to the existing subgrade (m), and $R^2$ is the coefficient of determination.

(2) Vertical displacement:

where $u_y$ is the vertical displacement (mm), and the other parameters are defined as above.

Fig. 5Displacement-distance fitting curves

a) Horizontal displacement

b) Vertical displacement

Current standards [15-16] stipulate allowable settlements of 30-50 mm for ballasted tracks at 300-350 km/h and displacement rates ≤ 3 mm/day for adjacent sections. Given the higher speeds of high-speed rail, stricter control is warranted. Simulation results show that horizontal and vertical displacements are of similar magnitude. Considering both directions, when the construction distance exceeds 60 m, the impact on the existing railway is negligible. At distances of 12-60 m, the influence is minor; at 1.8-12 m, the impact becomes significant; and when within 1.8 m, the effect is substantial. Practically, construction within 1.8 m requires reinforced retaining structures and real-time monitoring; 1.8-12 m needs standard monitoring with sheet pile support; beyond 12 m, routine procedures suffice.