Three-Dimensional Finite-Element Analysis of a Novel Pile–Anchor Intersecting Retaining System for Deep Excavations Adjacent to Existing Buildings

Abstract

To address the challenges of deep excavations adjacent to existing buildings—such as restricted anchor installation angles, construction disturbances, and excessive deformations—a novel pile–anchor intersecting retaining system is proposed. This study employs three-dimensional finite-element simulations via ABAQUS to analyze the mechanical behavior and deformation control performance of this new system. Four comparative models (cross-anchor with/without adjacent buildings, single-anchor with/without adjacent buildings) are established to investigate pile displacement, anchor axial force, and ground settlement during staged excavation. The results demonstrate that the proposed system exhibits superior mechanical behavior. It effectively reduces pile displacement and ground settlement, with reductions of up to 31.7% in final settlement compared to a conventional single-anchor system. The system’s design successfully avoids conflicts with adjacent foundations while providing high structural redundancy. In scenarios adjacent to existing buildings, the system performs safely, with maximum displacements well within code-specified limits. This study confirms that the pile–anchor intersecting system is a robust and practical solution for deep excavations in complex urban environments, offering enhanced safety and deformation control for projects near sensitive infrastructure.

1. Introduction

Deep excavations in densely built urban environments are increasingly common, yet they face a critical challenge when adjacent to existing buildings: the conflict between the need for inclined anchors to provide lateral restraint and the physical obstruction posed by neighboring foundations [1,2,3,4,5,6]. Consequently, optimizing support performance while controlling deformations—without compromising construction safety—has become a pressing priority in geotechnical engineering.

To address these challenges, extensive research has been conducted worldwide on the design, numerical simulation, and optimization of excavation support systems. Traditional approaches—such as limit-equilibrium and elastic foundation beam methods—are computationally efficient yet fail to capture the nonlinear soil–structure interaction along complex stress paths [7,8,9,10]. Recently, finite-element and finite-difference techniques (e.g., ABAQUS, FLAC3D), capable of realistically modeling construction sequences and dynamically tracking the mechanical response of support systems, have become indispensable tools for the analysis and design of deep excavations [11,12,13,14,15,16]. Mana and Clough [17] employed finite-element analyses to demonstrate that the maximum lateral wall displacement is governed by wall stiffness and basal restraint, proposing a simplified method for estimating wall movements. Addenbrooke and Dabee [18] examined over 30 case histories in stiff clay, using plane–strain nonlinear finite-element analyses under undrained conditions, and showed that displacement flexibility coefficients can be reliably applied to the design of multi-propped retaining structures.

Conventional pile–anchor retaining systems are the most widely used for deep excavations. However, when the excavation boundary lies immediately adjacent to existing buildings, two main conflicts arise. First, the entry and operation of large construction equipment inevitably disturb the surrounding ground, jeopardizing the safety of neighboring structures. Second, the presence of existing foundations prevents inclined anchors from being installed beyond a shallow angle, as their drilling path would otherwise intersect or undermine adjacent footings. Conventional pile–anchor systems often cannot install anchors at the required angles, compromising stability. While alternatives like double-row piles or internal bracing exist, they are frequently impractical due to space constraints, high material costs, or obstruction of excavation activities [19,20,21].

To address this gap, Jia Jin-qing introduced the “pile-anchor intersecting retaining system”, an innovative composite scheme that integrates front-row piles, mid-row low-angle inclined anchors, and rear-row vertical anchors. Monolithic casting of capping and connecting beams creates a spatial truss effect, delivering three-dimensional rigidity. This configuration markedly reduces excavation-induced deformations, eliminates the risk of anchors crossing adjacent foundations, and improves construction flexibility [22]. Nevertheless, its mechanical performance and deformation control efficacy have yet to be systematically validated through high-resolution numerical simulation, especially under complex geological conditions and staged excavation. In this study, a three-dimensional refined FE model is developed in Abaqus to investigate the behavior of the proposed system. Four comparative models are designed to quantify internal forces in the piles, wall deflections, anchor loads, and ground-surface settlements at each excavation stage. The new system is shown to significantly outperform the traditional conventional single-anchor system. Specifically, it achieves a 20.5% reduction in maximum pile displacement and a remarkable 31.7% reduction in ground settlement in the absence of adjacent buildings. Even under the additional load of an adjacent structure, the system maintains superior performance, reducing pile displacement and ground settlement by 17.9% compared to its conventional counterpart. These results demonstrate that the proposed system offers a highly effective solution for stringent deformation control in deep excavations adjacent to existing buildings. The findings offer practical guidance for future engineering applications of the system.

2. The Pile–Anchor Intersecting Retaining System

Figure 1 presents the structural configuration of the novel pile–anchor intersecting retaining system, highlighting the front-row piles, mid-row low-angle inclined anchors, rear-row vertical anchors, and monolithic capping and connecting beams that form a three-dimensional rigid frame. The low-angle mid-row anchors deliver the required horizontal restraint while remaining clear of neighboring foundations. Vertical rear anchors act as a simplified substitute for a second row of piles, eliminating the need for additional drilling close to existing structures and thereby minimizing construction disturbance. Staggering the mid- and rear-row anchors further reduces the potential for group-anchor interaction. Key features of the system are: [22]:

(1) Prestressed anchors work in concert with cantilever beams to generate a clockwise resisting couple, counteracting the overturning moment induced by active earth pressure on the piles (see Figure 1 for the system configuration).

(2) Monolithic casting of the front-pile capping beam, cantilever beams, and rear vertical-anchor waling beam forms a rigid frame, markedly enhancing global overturning resistance.

(3) The mid-row inclined prestressed anchors apply horizontal restraint to the piles via the cantilever and capping beams, significantly improving resistance to lateral deformation.

(4) The staggered arrangement of mid-row inclined and rear vertical anchors increases anchor spacing, effectively preventing group-anchor interaction.

3. Numerical Simulation

This paper established four groups of finite-element models corresponding to the support models through the finite-element software ABAQUS. For numerical simulation, foundation pit excavation involves various contact actions and boundary condition constraints. Large-scale foundation pit excavation simulation will greatly increase the computational cost, while too small a scale will affect the accuracy of the simulation results. Relevant research shows that outside the area four times the excavation depth of the foundation pit is the weak-influence area of foundation pit excavation [11,13,23]. This principle aligns with the guidance in the Abaqus Analysis User’s Manual [24], which recommends that model boundaries be placed sufficiently far from the region of interest to minimize their influence and to represent a semi-infinite medium. Based on this, the horizontal and vertical extents of the model are set to four times the excavation depth (4H = 40 m). The model dimensions are finalized as 60 m (length) × 4.4 m (width) × 40 m (height). The boundary conditions are applied as follows: the bottom surface is fully fixed (displacements in x, y, and z directions are constrained), simulating the rigid underlying rock layer; the left and right surfaces are roller boundaries, allowing vertical movement but restricting horizontal displacement in the x-direction, which prevents lateral flow of the soil mass; the top surface is a free surface with no displacement constraints. This configuration is a standard and well-validated approach in finite-element analysis of deep excavations, effectively isolating the excavation process as the primary source of disturbance. The influence of groundwater was not explicitly considered in the current analysis. This study focuses on the drained mechanical response of the pile–anchor intersecting system, assuming fully drained conditions for the soil. This simplification allows for a clearer investigation of the structural interaction and deformation control mechanisms of the new system under the primary loads of earth pressure and adjacent building surcharge, without the added complexity of hydro-mechanical coupling.

For the problem of foundation pit excavation, the Mohr–Coulomb (M-C) elastoplastic model is adopted to describe the soil behavior. While advanced constitutive models can capture more complex soil responses, the M-C model was selected for this study due to its widespread use, computational efficiency, and ability to capture the fundamental shear failure mechanism of soil. It is particularly suitable for a comparative study focused on the structural performance of the support system, as its parameters (cohesion and friction angle) are readily obtainable, and its behavior is well understood. The model assumes linear elasticity and a linear failure envelope in the stress space, which is a simplification of real soil behavior. However, for the monotonic, staged excavation loading analyzed in this work, the M-C model provides a robust and transparent framework for evaluating the relative performance of the proposed pile–anchor intersecting system [25]. Therefore, the soil constitutive model of this model adopts the Mohr–Coulomb elastoplastic model, with the element form of C3D8 and reduced integration hourglass control. The soil is characterized as a silty clay. Its mechanical properties are shown in

Table 1. Mechanical parameters of soil.

Density (kg/m3)	Elastic Modulus (MPa)	Poisson’s Ratio Static	Friction Angle (°)	Cohesion (KPa)
2700	70	0.3	35	24

. The influence of groundwater is not considered during the analysis. The supporting piles are concrete bored piles with a length of 17 m and a diameter of 800 mm, and the center-to-center spacing of the piles is 1.6 m. The element type is C3D8R. The top beam is simplified as a crown beam slab with dimensions of 5.4 m × 4.4 m. The structural elastic modulus E = 30 GPa, and the Poisson’s ratio is taken as 0.3. The row piles and the top beam are rigidly connected. The anchors are made of HRB400 steel bars with a diameter of 36 mm, with an elastic modulus E = 200 GPa, a Poisson’s ratio of 0.2, a density of 7850 kg/m³, and a linear expansion coefficient of 1 × 10⁻⁵. The element type is the T3D2 two-node linear three-dimensional truss element. For the tangential contact behavior, the penalty function is selected with a friction coefficient of 0.3, and for the normal contact behavior, hard contact is selected.

To ensure the reliability of the numerical results, a mesh sensitivity analysis was conducted prior to the final simulations. Multiple models with varying mesh densities were tested, focusing on the critical zones around the retaining piles and anchor installations. The analysis demonstrated that the key performance indicators—specifically, the maximum horizontal pile displacement and the peak ground settlement—converged to stable values when the maximum element size in these zones was refined to 0.8 m or smaller. The variation in results between the medium and fine mesh models was less than 5%. Based on this analysis, the mesh configuration used in the final models (as illustrated in Figure 2) is deemed sufficiently refined to provide accurate and mesh-independent solutions, thus verifying the numerical model.

The prestress of the anchor rod is applied by the cooling method. The principle is to cool the area where prestress is to be applied. According to the principle of thermal expansion and contraction, the materials adjacent to this area will be subjected to tensile forces in order to resist its shrinkage effect. For the anchor rod, when the temperature decreases, the prestressed anchor rod shrinks. Since the anchor rod is connected to the soil through the embed command, and the top is anchored to the beam–slab through the tie command, the shrinkage strain of the prestressed tendon can be transmitted to the materials closely connected to it. It should be noted that when applying the temperature effect, the entire regional structure range cannot be selected. The anchor rod should be cooled separately. This is because cooling the overall structure will not generate a shrinkage strain difference. The prestress value designed in this paper is 120 kN. The cooling method is used to apply prestress to the anchor rod, and the equivalent temperature obtained is −59 °C. The calculation formula is as follows:

$$\mathsf{\Delta }T=-{\displaystyle \frac{F}{\alpha EA}}$$

(1)

$\mathsf{\Delta }T$—Temperature difference of the bolt °C;

$F$—Bolt prestress N;

$\alpha$—Coefficient of linear expansion of anchor bolt. This value is set in the material properties 1⁻⁵;

$E$—Elastic modulus of anchor bolt Pa;

$A$—Cross-sectional area of the anchor bolt m².

It is important to note that the coefficient of linear expansion used here is a fictitious parameter chosen for numerical convenience. It is not the actual thermal expansion coefficient of steel. This technique provides an accurate and efficient way to apply a known, uniform prestress force. However, it is a simplification that assumes a perfectly uniform stress distribution along the anchor’s free length, which may not capture localized stress concentrations near the anchor head or in the bond zone. Despite this simplification, the method is widely accepted and provides a robust representation of the overall prestressing effect for the purpose of this comparative study.

Model Grouping and Excavation Process

To quantitatively evaluate the performance enhancement of the novel pile–anchor intersecting system, a direct comparative analysis was conducted against a conventional single-anchor pile–anchor support structure, which represents the most common traditional solution in deep excavation practice. The test models in this paper are divided into four groups, namely, no-building cross-anchor type, no-building single-anchor type, with-building cross-anchor type, and with-building single-anchor type. The schematic diagram of the with-building cross-anchor structure is shown in Figure 2. For the “with-building single-anchor type”, the diagonal anchors in the middle row are removed based on the “with-building cross-anchor type”. To simulate the influence of an adjacent existing building, a uniformly distributed surface load of 60 kPa is applied to the ground surface at a horizontal distance of 3 m from the retaining structure, over an area of 4.4 m × 10 m. This approach is a widely used simplification in geotechnical numerical modeling for deep excavations adjacent to buildings. While this method does not explicitly model the detailed interaction between the building’s foundation (e.g., footings or piles) and the soil, it provides a consistent and well-defined external surcharge that effectively represents the presence of the building’s weight. This simplification allows for a clear comparative analysis between the two support systems under a controlled loading condition, which is the primary focus of this study. The use of a uniform pressure avoids the added complexity of modeling specific foundation types and their interface behavior, ensuring the analysis remains focused on the performance of the retaining system itself. The “no-building cross-anchor type” and “no-building single-anchor type”, respectively, remove the influence of the building load based on the “with-building cross-anchor type” and “with-building single-anchor type”. The following steps are used to simulate the above four working conditions:

Soil body modeling, followed by in situ stress equilibrium;

Install pile row support and anchors;

kw1: Excavate to −1.4 m;

kw2: Excavate to −4.4 m;

kw3: Excavate to −7.4 m;

kw4: Excavate to the bottom of the pit at −10.4 m.

Record the impact of layered excavation of the foundation pit under different working conditions on the support structure, including the changes in the horizontal displacement of the pile top, the axial force of the prestressed anchor rod, and the bending moment of the row piles, as well as the settlement changes in the ground outside the pit and around the buildings.

Figure 2. Schematic diagram of the anchor structure under the adjacent building.

Figure 2. Schematic diagram of the anchor structure under the adjacent building.

4. Numerical Simulation Results Analysis

Figure 3 shows the soil displacement nephogram after the in situ stress balance. The magnitude of soil displacement is below 10⁻¹⁴, and the soil displacement is almost negligible. It is far less than the evaluation standard of 1 × 10⁻⁴, so it can be used as the initial state of foundation pit excavation.

4.1. Analysis of Horizontal Displacement of the Body

4.1.1. Horizontal Displacements in Non-Building Group

Figure 4 shows the deformation nephogram of the pile body displacement (U1) of the row pile-and-anchor cross structure in the no-building group (with a magnification factor of 200). In the Figure, “+” represents the displacement towards the excavation face of the foundation pit, and “−” represents the displacement towards the opposite side of the excavation face of the foundation pit. From the horizontal displacement nephogram of the pile body, it can be found that the deformation of the pile body is relatively small in the initial stage of foundation pit excavation, and the maximum displacement occurs in the top area of the pile, with a horizontal movement of about 0.69 mm towards the outside of the pit. As the excavation depth continuously increases, the deformation of the pile body gradually becomes obvious. Negative displacement in the direction towards the inside of the pit begins to appear at the pile top, and the area of the maximum negative displacement continuously moves downward with the excavation of the foundation pit. The displacement of the pile body shows a “convex deflection pattern” deformation.

Figure 5 shows the variation diagram of the horizontal displacement of the pile shaft under various working conditions in the no-building group. As can be seen from Figure 5a, in the initial stage of foundation pit excavation, the pile shaft produces displacement deformation towards the outside of the pit, but the deformation amount is very small, and the horizontal displacement of the pile top is 0.464 mm. This is because the pile body is affected by the horizontal pulling force of the anchor rod at the pile top. When the excavation depth is small, the pile top is horizontally constrained and produces a small displacement towards the outside of the pit. With the continuous increase in the excavation depth, the earth pressure behind the pile continues to increase, the horizontal displacement of the pile top continues to grow towards the outside of the pit, and the pile shaft displaces towards the excavation surface side. The maximum displacement amounts under the various working conditions are −2.36 mm, −4.84 mm, and −7.38 mm, respectively. It can be found that the displacement of the pile shaft is positively correlated with the depth of the foundation pit. At the same time, it can be seen from Figure 5a that there is an inflection point in the pile body (i.e., the displacement of the pile body changes from positive to negative). Near the excavation surface under various working conditions, there is a small displacement of the pile body towards the outside of the pit. The maximum displacements outside the pit for kw2–kw4 are 1.02 mm, 0.93 mm, and 0.44 mm, respectively. The displacement of the pile body outside the pit gradually decreases as the excavation depth of the foundation pit increases, and the position of the displacement inflection point is approximately at the excavation surface of each stage. This is because the deformation of the pile top is controlled by the horizontal tension of the anchor rod. The soil behind the pile has a tendency to move towards the inside of the pit due to the unloading of the soil in front of the pile, resulting in an increase in the active earth pressure. A resistance effect is formed among the horizontal tension of the anchor rod, the earth pressure behind the pile, and the reaction force of the soil in front of the pile. When the foundation pit is continuously excavated deeper, the reaction force of the soil in front of the pile is difficult to resist the rapidly increasing earth pressure behind the pile. Therefore, the displacement of the inflection point outside the pit gradually decreases. After the excavation of the foundation pit is completed, the pile body shows an obvious “bulging” deformation.

It can be seen from Figure 5b that after the excavation of the first layer is completed, the displacement and deformation of the pile top towards the pit occur. As the excavation of the foundation pit continues, the horizontal displacement of the pile body increases significantly, and the maximum reverse bending displacement of the pile body is 0.43 mm. The whole pile body deforms towards the pit, and the maximum horizontal displacements of the pile body are −0.53 mm, −2.61 mm, −5.31 mm, and −8.89 mm, respectively. By comparing the cross-anchored type and the single-anchored type, it can be found that after the excavation is completed, the maximum horizontal displacement of the pile body of the cross-anchored type is reduced by 20.5% compared with that of the single-anchored type, indicating that the cross-anchored type has a stronger ability to restrain the deformation of the pile body than the single-anchored type, and the control effect is better.

Figure 6 shows the stress condition of the pile body in the typical anchor-intersection structure. Taking the displacement and deformation diagram of the pile body after excavation as an example, the forces acting on the intersecting pile body, from top to bottom, are as follows: the horizontal right-ward pulling force of the anchor support system, the horizontal left-ward thrust of the soil behind the pile, and the horizontal right-ward reaction force of the soil below the excavation surface in front of the pile. First, since the top of the anchor rod of the row/pile–anchor-intersection support structure is anchored on the top beam, the pile top will continuously be subject to a horizontal right-ward constraint. However, as the depth of the foundation pit increases, the earth pressure behind the pile continuously increases, and this restraining effect will gradually decrease. Second, due to the horizontal constraint at the pile top, in order to resist the increasing stress of the soil behind the pile caused by excavation unloading, the pile body bears most of the earth pressure behind the pile. At this time, it is similar to the stress mode of beam bending, and the pile body will produce a deformation characteristic of bending towards the pit, forming a “bulging-belly shape”. Finally, the soil below the excavation surface in front of the pile, as the embedded soil of the pile, will provide a horizontal right-ward reaction force when it is subjected to the deformation of the pile bending towards the pit. Therefore, a small inflection point will appear at the excavation surface in front of the pile for the deformation of the pile body. For the single-anchor type, due to the lack of the middle-row inclined anchor rod, the horizontal resistance of the entire support system is insufficient, and the soil pressure behind the pile dominates, exacerbating the development of the pile body’s deformation towards the pit. Ultimately, the deformation of the pile body is greater than that of the anchor-intersection type.

4.1.2. Pile Top Horizontal Displacement

Figure 7 is a comparison diagram of the horizontal displacement of the pile body after the excavation of four groups of model tests. It can be seen from the Figure 7 that the displacement of the pile body basically shows a “convex deflection pattern” shape. The deformation degrees of the two groups of models with buildings are greater than those of the groups without buildings. The maximum horizontal displacement of the cross-anchor type with buildings increases by 26.8% compared with that without buildings, and the maximum horizontal displacement of the single-anchor type with buildings increases by 28.9% compared with that without buildings. Moreover, the horizontal displacement of the pile top of the groups with buildings is greater than that of the groups without buildings. This indicates that under the same conditions, the existence of buildings will exacerbate the deformation degree of the pile body.

From the comparison between the cross-anchored and single-anchored groups, it can be found that the maximum horizontal displacement at the pile top of the cross-anchored structure with and without building conditions is reduced by 87.3% and 83.2%, respectively, compared with the single-anchored structure. This indicates that the cross-anchored structure has a better ability to control deformation than the single-anchored structure. This is because the restraint force of the system after removing the middle row of anchor rods is insufficient to resist the gradually increasing earth pressure behind the pile. Especially, the existence of the building further compresses the soil behind the pile, increasing the soil stress, and the active earth pressure behind the pile will further increase. However, due to the existence of the middle row of anchor rods, the cross-anchored structure provides a greater horizontal resistance to the deformation of the supporting structure. The supporting system after removing the middle row of anchor rods can only control the trend of the pile body moving towards the inside of the foundation pit to a certain extent. The absence of the middle row of anchor rods greatly weakens its ability to resist deformation. The maximum horizontal displacements of the two groups of the single-anchored structure are both greater than those of the two groups of the cross-anchored structure, which also demonstrates the strong restraint ability of the cross-anchored structure in controlling the deformation of the foundation pit.

According to the control limits of the retaining structure in the “Technical Code for Retaining and Protection of Building Foundation Excavations (JGJ 120-2012)” [26], for a second-level foundation pit, the maximum lateral displacement of the support structure is 0.3%H, where H is the depth of the foundation pit excavation. According to the above requirements, the maximum lateral displacement limit of the support structure of this model is 31.2 m. Among the four groups of models, the maximum lateral displacement occurs in the single-anchored type with a building, and the maximum lateral displacement is 12.5 mm, which is 38.9% of the limit value of the foundation pit, meeting the safety requirements of the foundation pit. The maximum lateral displacement of the cross-anchor type is only 29.2% of the limit value of the foundation pit. This indicates that the row pile–anchor cross structure can meet the safety requirements for the use of the foundation pit, and its performance in controlling deformation is better than that of the single-anchored structure without the middle row of inclined anchor rods.

In summary, since the row pile–anchor intersection support is a three-dimensional rigid structure, it theoretically has strong anti-overturning ability. The anchors in the middle and rear rows transmit the anchoring force of the deep soil layer to the top along the axial direction. The beam at the top is equivalent to the integral casting of the capping beam and the connecting beam, forming a relatively large stiffness to resist overall deformation. Moreover, as the excavation depth of the foundation pit increases, the earth pressure behind the piles continuously pushes the pile body to develop towards the inside of the pit. The top of the anchor, due to its connection with the beam, reduces the displacement of the pile top towards the inside of the pit. This indicates that the row pile–anchor intersection support system has a good ability to control deformation and plays a significant role in preventing the inward inclination of the retaining piles.

4.1.3. The Internal Force of Inclined Bolts

The overall force-bearing process of the prestressed anchor is that the free-length section of the anchor is first affected by the prestress, and then the force is transmitted axially to the fixed-length section of the anchor. In this model, the prestress is applied by the cooling method. A prestress temperature field of −59 °C is applied to the rod body to simulate the application of a prestress of 120 kN. The free-length section transmits the tensile force to the fixed-length section. The anchor reinforcement exerts the uplift resistance on the entire section of the anchor through the friction with the deep soil. The anchor used in this model is HRB400 steel bar, with a design value of tensile strength of 360 N/mm² and a standard value of yield strength of 400 N/mm². To ensure the normal operation of the steel bar without damage, the design value of tensile strength is taken as the limit. Converted into the maximum stress, it is 3.6 × 10⁸ Pa, and the maximum axial force is 366.4 kN. That is, when the stress or axial force of the anchor exceeds this value, the anchor is considered to be damaged.

Figure 8 shows the nephograms of the stress changes in the inclined anchors in two sets of models with cross-anchored type without buildings. It can be seen from the Figure 8 that during the excavation of each working condition of the foundation pit, the stress distribution in the free section of the anchors is basically the same, and the area with the maximum stress is within the free section range. According to the force characteristics of prestressed anchors, the force transfer of the anchors is from the free section to the anchored section. The tension on the free section is basically equal, and the anchored section transfers the force of the free section to the deeper soil layer under the shear action with the soil. It can be seen from the nephograms that the maximum stress of the inclined anchors (1.667 × 10⁸ Pa) is significantly lower than the ultimate tensile strength of the HRB400 steel bar (3.6 × 10⁸ Pa), meeting the material strength requirements. More importantly, this maximum stress is well within the allowable stress limit prescribed in design practice. For HRB400 steel, the allowable stress is commonly taken as 70% of its yield strength (400 MPa), resulting in an allowable value of 280 MPa (2.8 × 10⁸ Pa). The observed maximum stress of 166.7 MPa represents only 59.5% of this allowable limit, indicating a substantial safety factor and confirming the safety and adequacy of the anchor design under the simulated conditions. This is due to the soil unloading caused by the excavation of the soil in front of the pile. The soil pressure behind the pile continues to increase, and the soil generates lateral deformation towards the inside of the pit. The support structure displaces towards the inside of the pit under the action of the soil pressure, resulting in the continuous increase in the anchor tension. This also shows that the axial force of the anchor is positively correlated with the excavation depth of the foundation pit.

Figure 9 shows the nephograms of the stress changes in the inclined bolts in two groups of models with building-anchored intersection. It can be seen from the Figure 9 that after the excavation of the foundation pit, the stress of the bolts increased by 37.1% compared with that of kw1. In the group with buildings, the maximum bolt stresses at the initial stage of excavation (kw1) and at the end of excavation (kw4) increased by 9.6% and 15.7%, respectively, compared with those in the group without buildings. This indicates that when there are buildings, the axial force of the bolts during the excavation of the foundation pit is greater than that in the group without buildings. From the stress nephogram, it can be seen that the stress in the front area of the free section of the bolts in the kw4 stage is greater, but the axial force in the free section is generally approximately equal. This shows that when there is a large load outside the pit, there is a certain stress concentration in the top area of the bolts, and there is a slight stress loss along the length direction of the free section.

4.1.4. The Internal Force of Vertical Bolts

Figure 10 shows the vertical prestressed anchor stress nephograms after the excavation of each group of foundation pits. Vertical anchors exist in all four groups of models. Therefore, this paper compares the anchor stress nephograms after the excavation of the foundation pits of each model to obtain the stress distribution law of the vertical anchors. As can be seen from Figure 10, the stress distribution of the anchors is mainly concentrated near the free section. It can also be seen from the nephograms that the maximum stress of the vertical anchors does not exceed 3.6 × 108 Pa, meeting the strength requirements.

Figure 10a,b show the vertical bolt stress nephograms in the cross-anchored structure without and with buildings, respectively. It can be seen from the Figure 10 that after the excavation of the foundation pit, the maximum stress of the vertical bolts of kw4 decreases by 51.5% compared with that of the inclined bolts in the case without buildings, and the maximum stress of the vertical bolts of kw4 decreases by 21.1% compared with that of the inclined bolts in the case with buildings. This indicates that the stress of the inclined bolts is greater than that of the vertical bolts during the excavation of the foundation pit. Under the condition of the existence of buildings, the maximum stress of the vertical bolts increases by 88.3% compared with that without buildings. This shows that the increase in the load outside the pit will increase the stress of the vertical bolts.

Figure 10 shows the stress nephograms of the vertical anchors in the single-anchor structure without and with buildings. It can be seen from the Figure 10 that, with and without buildings, the maximum stress of the vertical anchors after removing the middle row of anchors increases by 19.3% and 88.6%, respectively, compared with that of the vertical row of anchors in the anchor-intersection structure. This indicates that the rear-row vertical anchors bear more axial tension when acting alone. Especially, when there are buildings, the axial force of the vertical anchors when acting alone is only 6.1% less than the maximum stress of the inclined anchors in the anchor-intersection structure, which also shows that the existence of buildings does significantly increase the axial force of the anchors.

Through the above comparative analysis of the stress of the anchor rod, it can be found that the existence of the existing building will increase the maximum stress of the anchor rod. The working area of the anchor rod is concentrated in the free section, and the axial force distribution on the free section is relatively uniform. However, there will be stress loss at the interface between the free section and the anchorage section, resulting in the force on the free section not being able to be completely transmitted to the deep soil behind the pile. This is because the anchor rod in the free section is first subjected to the tension brought about by the deformation of the support structure. When the anchorage section is stressed, the force is transmitted to a deeper area through the friction with the surrounding soil. However, in the actual situation, with the deformation of the foundation pit, the stress of the anchored soil is also continuously changing. The displacement of the soil often leads to the problem of insufficient grip force in the anchorage section relying on the friction between the rod and the soil, so stress loss in the anchorage section often occurs.

For the stress characteristics of vertical bolts, when the support system shows a deformation trend of tilting towards the inside of the pit, the bolts in the rear row are subjected to more vertical pulling forces, which is not conducive to the force transfer of the overall support structure. Prestressed bolts have mechanical characteristics such as providing active resistance and being able to transfer the tension of the structure to the deep stable stratum.

Through the above comparison, it can be found that the placement angle of the anchor bolt is closely related to its force transmission path. Although the vertically placed anchor bolt causes less disturbance to the soil behind the support and can avoid the damage to the surrounding existing buildings, the vertical anchor bolt is more prone to pull-out failure. When the foundation pit deforms, the front row of piles gradually generates an inward inclined displacement. At this time, the support system needs to provide a large horizontal pulling force. Therefore, the existence of the middle row of anchor bolts is indispensable. A complete support system can make up for the deficiency of horizontal resistance, and the staggered layout system can also avoid the occurrence of the group anchor effect.

4.2. Influence of Bolt Prestress

To discuss the influence of the prestress magnitude of the anchor rod system in the row pile–anchor alternating structure on the support effect of the retaining pile, the influence of the prestress magnitudes of 100 kN, 150 kN, 175 kN, and 200 kN on the displacement and bending moment of the retaining pile body is shown in Figure 11.

It can be found from Figure 11 that within a certain range, increasing the prestress of the anchor bolt can effectively reduce the horizontal displacement and bending moment of the pile body. When the prestress of the anchor bolt is 100 KN, the maximum displacement and bending moment of the pile body are 10.32 mm and 302.88 KN, respectively. When the prestress of the anchor bolt is 200 KN, the maximum displacement and bending moment of the pile body are 6.44 mm and 141.26 KN, respectively. The maximum displacement and maximum bending moment of the pile body decreased by approximately 37.6% and 53%, respectively. As the excavation depth continues to increase, the soil pressure on the pile body also increases. The anchor bolt system provides a horizontal component of force for the support structure, assisting the pile body to jointly resist the deformation caused by soil unloading. The role of the prestressed anchor bolt can effectively reduce the bending moment of the pile body. However, the increase in prestress should be within an appropriate range. In practical engineering, through the interaction between the steel bar of the anchor bolt and the grouting body, and through the force transfer between the grouting body and the surrounding soil, the anchor bolt is firmly anchored in the soil to form a prestressed anchor bolt with a supporting effect. Therefore, when increasing the prestress, the construction site environment and soil conditions should be comprehensively considered, and the magnitude of the applied prestress should be reasonably selected.

4.3. Ground Settlement Outside the Pit

According to the control limits of foundation pit excavation deformation in the “Technical Code for Retaining and Protection of Building Foundation Pits (JGJ 120-2012) [26]”, for a second-level foundation pit, the maximum settlement of the ground outside the pit is 0.25%H. Here, H is the depth of the foundation pit excavation. According to the above requirements, the ground settlement limit of this model is 26 mm. When the maximum settlement value of the foundation pit is greater than the limit, the foundation pit is considered to be damaged.

4.3.1. Ground Surface Settlement Outside the Pit in the Group Without Buildings

Figure 12 shows the variation of ground surface settlement outside the foundation pit for the single-anchor type without buildings and the anchor-intersection type without buildings. From Figure 12a, it can be seen that the maximum settlement values of each working condition in the single-anchor type are 4.44 mm, 6.19 mm, 8.08 mm, and 10.64 mm, respectively. Compared with the kw1 working condition, the settlement of the remaining working conditions increases by 39.4%, 81.5%, and 139.6%, respectively. It can be found that with the increase in the excavation depth, the settlement deformation of the soil outside the foundation pit gradually increases, and the overall growth amount continues to increase. From Figure 12b, it can be seen that the maximum settlement deformation values of each working condition in the anchor-intersection type are 3.79 mm, 5.02 mm, 6.49 mm, and 8.54 mm, respectively. Compared with the kw1 working condition, the settlement of the remaining working conditions increases by 32.4%, 71.2%, and 125.3%, respectively. After the excavation is completed, the maximum ground surface settlement value of the anchor-intersection type is 31.7% lower than that of the single-anchor type. Through comparison, it can be seen that the growth amount of settlement deformation of each working condition in the anchor-intersection type is less than that in the single-anchor type during the foundation pit excavation process. This indicates that the ground surface settlement outside the foundation pit is well controlled under the action of the anchor-intersection type, and its advantage in controlling the settlement deformation of the foundation pit is better than that of the single-anchor type.

It can be seen from Figure 12b that during the initial stage of foundation pit excavation, there is a short-term heave deformation in the anchor-intersection type. The area is within 2 m near the outside of the pit, and the maximum heave deformation is 2.91 mm. As the excavation depth increases, it transforms into settlement. This indicates that during the initial stage of foundation pit excavation, the combined action of inclined and vertical anchors will cause extrusion deformation of the soil around the foundation pit. However, as the foundation pit is excavated, the earth pressure behind the pile increases, the soil stress is redistributed, and the influence of soil settlement deformation is greater. Given its small magnitude and transient nature, this uplift is unlikely to have any detrimental impact on the serviceability of adjacent pavements or structures. The primary deformation of concern, which is the final ground settlement, is significantly reduced by the proposed system.

Overall, for the two groups of models without buildings, the maximum settlement of the soil body occurs at a position about 4–6 m outside the foundation pit, which is approximately 0.4–0.6 times the excavation depth. “Settlement troughs” appear in the surface settlement. As the distance from the pit edge increases, the soil settlement gradually decreases. With the continuous increase in the excavation depth of the foundation pit, the range of the “settlement trough” gradually expands. This indicates that the excavation of the foundation pit will intensify the settlement effect of the ground outside the pit, and this effect will weaken as the distance from the pit edge increases. Under the combined action of the middle and rear row anchor rods, the anchor-intersection type shows an upward heaving phenomenon of the soil body in the initial stage of foundation pit excavation, but the deformation amount decreases rapidly and gradually turns into a settlement trend. This shows that the deformation trend of the ground is mainly settlement. Only in the initial stage of foundation pit excavation, due to the extrusion of the support structure, local instantaneous heaving occurs. With the stress redistribution of the soil body, the surface settlement caused by soil unloading will become dominant. Through comprehensive analysis, in the case of no buildings, the ability of the anchor-intersection type to control the deformation of the foundation pit is better than that of the single-anchor type.

4.3.2. Ground Settlement Outside the Pit No Buildings

Figure 13 shows the changes in the ground settlement outside the pit for the single-anchor type with buildings and the anchor-intersection type with buildings. As can be seen from Figure 13a, the maximum settlement values of each working condition in the single-anchor type are 7.55 mm, 10.94 mm, 13.85 mm, and 16.99 mm, respectively. Compared with the kw1 working condition, the settlement of the remaining working conditions increases by 44.9%, 83.4%, and 125.9%, respectively. Similar to the group without buildings, with the increase in the excavation depth, the settlement deformation of the soil outside the pit also continues to increase. The difference is that the maximum ground settlement affected by each level of excavation in the group with buildings is greater than that in the group without buildings. As can be seen from Figure 13b, the maximum settlement deformation values of each working condition in the anchor-intersection type are 6.98 mm, 9.46 mm, 11.44 mm, and 13.94 mm, respectively. Compared with the kw1 working condition, the settlement of the remaining working conditions increases by 35.5%, 63.9%, and 99.7%, respectively. After the excavation is completed, the maximum ground settlement value of the anchor-intersection type is 17.9% lower than that of the single-anchor type.

It can be seen from Figure 13b that during the initial stage of foundation pit excavation, in the kw1–kw2 stage, the anchor-intersection type shows a brief uplift. After the excavation of the first layer of soil is completed, the maximum uplift amount is 2.23 mm. Subsequently, as the excavation depth of the foundation pit increases, the soil outside the pit turns into a settlement situation. This is because the “soil-squeezing effect” of the support structure in the initial stage of excavation causes a brief uplift of the soil outside the pit. With the excavation of the foundation pit and the existence of the load of the buildings behind, the stress of the soil behind the piles will increase significantly, resulting in the gradual weakening and disappearance of this effect.

Overall, the “settlement troughs” appeared in the surface settlement changes outside the pits of the two groups of models. The location where the maximum settlement value occurred was around 6–8 m outside the pit, which was approximately 0.6–0.8 times the excavation depth. This is because simulated existing buildings were arranged outside the foundation pit, and the maximum surface settlement occurred around the buildings. As the excavation depth of the foundation pit increased, the surface settlement continued to increase. In the case of the existence of existing buildings, the maximum settlement of the single-anchor type increased by 29.7% compared with the cross-anchor type. The reason is that the existence of the buildings outside the pit generated additional stress on the soil. For the single-anchor type, due to the lack of the middle row of inclined anchors, the control effect on the deformation of the foundation pit weakened, the pores of the soil behind the pile were compressed, and the earth pressure increased, resulting in larger surface settlement. This indicates that the existence of existing buildings will exacerbate the surface settlement outside the foundation pit, and the cross-anchor system with the combined action of the middle row and the rear row of anchors can reduce the maximum surface settlement.

5. Conclusions

Numerical simulations using the ABAQUS software were conducted on four different models to evaluate the performance of the proposed pile–anchor intersecting retaining system. Based on the analysis, the following conclusions are drawn:

(1) As the excavation depth continuously increases, the deformation of the pile body gradually becomes obvious. Negative displacement in the direction of the pit begins to appear at the pile top, and the area of the maximum negative displacement continuously moves downward with the excavation of the foundation pit. The displacement of the pile body shows a “bulging-belly” deformation. The existence of buildings will exacerbate the degree of pile body deformation. The maximum horizontal displacement of the anchor-intersection type with buildings is 26.8% higher than that without buildings. The anchor-intersection structure has better deformation control ability than the single-anchor structure. The maximum horizontal displacements at the pile top of the anchor-intersection structures with and without buildings are reduced by 87.2% and 82.2%, respectively, compared with the single-anchor structure.

(2) During the excavation of the foundation pit, the axial force distribution of the free section of the anchor rod is basically the same. The area with the maximum stress is within the free-section range. The force transfer of the anchor rod is from the free section to the anchorage section, and there is a loss of axial force in the anchorage section. The magnitude of the axial force of the anchor rod is positively correlated with the excavation depth of the foundation pit. The existence of buildings will increase the maximum axial force of the anchor rod. Under the condition of the existence of buildings, the maximum axial force of the vertical anchor rod is 88.3% greater than that without buildings. The rear-row vertical anchor rod bears more force when acting alone, but the force of the inclined anchor rod is greater than that of the vertical anchor rod during the excavation of the foundation pit. Appropriately increasing the prestress of the anchor rod can effectively control the deformation of the pile body.

(3) The deformation trend of the ground surface is mainly settlement. In the initial stage of foundation pit excavation, the extrusion effect of the support structure causes local instantaneous uplift of the soil. With the stress redistribution of the soil, the ground-surface settlement caused by soil unloading becomes dominant. The ground-surface settlement shows a “funnel” shape. The areas of maximum settlement appear near 6–8 m and 4–6 m outside the support with and without buildings, respectively, that is, at 0.6–0.8 times and 0.4–0.6 times the excavation depth. The existence of buildings will exacerbate the settlement change in the ground surface, and a “settlement trough” will appear near the building. The trough surface gradually increases with the excavation of the foundation pit.

(4) In the case of being close to an existing building, the maximum lateral displacement value of the pile body of the row/pile–anchor-intersection structure is 9.32 mm, which is 29.8% of the lateral displacement limit of the foundation pit; the maximum settlement value outside the pit is 13.94 mm, which is 53.6% of the settlement limit of the foundation pit. The anchor-intersection structure can meet the safety requirements of the foundation pit in the case of an existing building and has good practicality.

The pile–anchor intersecting system proposed in this study offers a structurally efficient and space-saving solution for deep excavations in densely built urban environments, particularly those adjacent to existing buildings. By replacing the conventional second row of piles with vertical rear anchors and incorporating 15° inclined front anchors, the system achieves comparable structural rigidity while significantly reducing material consumption and the construction footprint. This design not only minimizes ground disturbance and avoids potential conflicts with neighboring foundations but also maintains constructability using standard drilling equipment and established construction sequences—pile installation, staged excavation, anchor installation with prestressing, and monolithic casting of capping beams. Numerical results demonstrate that the system effectively suppresses wall deflections and ground settlements, providing a high safety margin for adjacent infrastructure.

Nevertheless, the system’s performance is contingent upon favorable geological conditions, particularly, the presence of a competent stratum to ensure sufficient anchor pull-out resistance. The influence of groundwater, though not explicitly modeled, must be carefully evaluated in practice, as high water tables can compromise soil strength and stability. Moreover, while the numerical simulations show promising mechanical behavior, the current study is based on deterministic modeling without physical validation. Future work will include centrifuge model testing to experimentally verify the system’s performance and a parametric study to assess its sensitivity to variations in key soil parameters, such as cohesion, friction angle, and elastic modulus. These efforts are essential to bridge the gap between numerical prediction and practical application, supporting the broader adoption of this innovative system in real-world engineering projects.