Deformation Response of a Pipeline to Nearby Deep Foundation Pit Excavation: Numerical Simulations and Field Tests

Abstract

Deep foundation pit excavation (DFPE) for metro stations in soft layers can cause strata deformation, impairing the safety and performance of nearby existing structures. This paper aims to explore the impact of DFPE on an adjacent oil pipeline using numerical simulations and field tests. Using the project of Xinxingcun Station of Tianjin Metro Line 4 in China as a case study, two approaches of layered excavation (LE) and bench excavation (BE) are proposed for DFPE. The deformation induced by the two approaches was analyzed using the FE method. The maximum vertical settlement of the oil pipeline caused by LE was 8.1 mm, compared with the maximum vertical settlement of 12.5 mm caused by BE. In practice, LE was selected as the construction scheme for the DFPE in Xinxingcun Station. Some field measurement work on pipeline deformation was conducted, and the results were in good agreement with the numerical simulation. Lastly, the change pattern in the pipeline-soil gap was studied. The angle of the position with peak gap width increased from 25° to 110°.

1. Introduction

Metro transit has greatly developed around the world as a green and environmentally friendly means of underground transportation. However, most metro transit systems were built in urban centers, where the large number of underground pipelines increases the difficulty of metro transit construction [1]. In particular, the open-cut method has prevailed in the construction of metro stations, exerting a significant impact on the surrounding environment. If there are oil pipelines nearby, foundation pit excavation becomes significantly more difficult. Currently, the impact of deep foundation pit excavation (DFPE) on adjacent structures has become a key issue to be solved in metro transit construction [2].

Many studies on the impact of DFPE on pipelines have been conducted. Peck [3] proposed an empirical formula to predict the ground surface settlement after analyzing a large number of ground surface settlement data. Crofts et al. [4] presented an estimation method for the horizontal displacement of underground pipelines caused by nearby foundation pit excavation. Drawing on Winkler’s theory, Li et al. [5] created an equation expressing how foundation pit excavation affects the vertical and horizontal displacements of a pipeline. Klar et al. [6] studied the soil-pipe-tunnel interaction using two calculation methods. It was found that, when the relative material stiffness of the pipe and soil is small, the pipe does not behave as a beam. Zhang et al. [7] provided a continuous elastic analysis method in finite difference form aimed at simulating the responses of both continuous pipelines and joined pipelines subjected to tunnel-induced soil movement in multilayer soils. Jiang et al. [8] analyzed the empirical formula regarding soil deformation on the side of foundation pits, established the differential equation of elastic foundation beams, and derived the analytical solution of pipeline deformation and internal force. Zhang et al. [9] proposed a simplified method to study the mechanical behavior of buried pipelines induced by foundation pit excavation using Winkler’s foundation model. Tan et al. [10] put forward an empirical formula for estimating pipeline displacement caused by the excavation of adjacent foundation pits based on a regression analysis of field-measured data. These studies provided theoretical and empirical results. In addition, some scholars studied the monitoring of actual foundation pit excavation projects [11,12], and analyzed the settlement in such projects [13,14].

With the development of computer technology, there has been extensive research work on the numerical analysis of how foundation pit and tunnel excavation affects adjacent buildings since the beginning of the 21th century [15,16,17]. Doležalová [18] studied the impact of deep excavation on an underlying tunnel through finite element modeling, and predicted and analyzed the deformation and stress changes of the tunnel during the excavation process. Yang et al. [19] used the finite element model to study the relationship between foundation pit excavation depth and pipeline displacement. Shi et al. [20] studied the response of a pipe subjected to the explosion shock of a nearby pipe through finite element analysis and discussed how the intersection angle of a burst pipe centerline and the adjacent pipe affects the bursting effect. Al-Khazaali et al. [21] numerically analyzed the behaviors of buried rigid and flexible pipelines using the effective stress analysis method. It was found that trenching in unsaturated soil contributes to deformations of the soil surrounding the buried pipelines and results in lower internal forces. Sun et al. [22] analyzed how the pressure difference between soils inside and outside a foundation pit affects the load and lateral deformation of the supporting structure using 3D and 2D finite element methods. Zhang et al. [23] established a 3D model for pipelines and foundation pits, and studied the pattern of pipeline deformations at different foundation pit excavation parameters. Ye et al. [24] analyzed the influence of foundation pit excavation on the deformation and internal force of adjacent subway tunnels through finite element simulation based on engineering examples, and optimized the foundation pit structure accordingly.

In summary, many achievements have been made in the research on the deformation of adjacent existing pipelines and soil caused by foundation pit and tunnel excavation, which provides theoretical support for solving practical engineering problems. However, most existing research and simulations were based on the assumption that the pipeline is in full contact with soil during deformation [25,26,27]. In fact, buried pipelines interact with the surrounding soil. They cannot be in full contact with the soil, especially in soft soil areas, where the soil is easily deformed in a manner that makes pipelines difficult to achieve complete contact with the soil. Therefore, there is a need to further examine how the excavation of deep foundation pits in soft soil affects nearby pipelines.

To this end, this paper investigates the impact of foundation pit excavation on nearby aviation oil pipelines in the aviation oil pipeline construction project for Xinxingcun Station of Tianjin Metro Line 4 in China. With the help of finite element software Midas, the paper simulates layered excavation and bench excavation, and carries out a comparative analysis of how the two construction methods influence the vertical settlement of adjacent pipelines and the deformation of nearby soil. On that basis, it discusses the non-coordinated deformation mechanism for the pipelines and soil in the excavation process of the deep foundation pits for the stations, and clarifies the pattern of changes in the pipeline-soil gaps. The patterns of the deformation of the soil and pipelines around the excavated foundation pit are observed in the field and verified by comparing the numerical simulation results with the actual results. Lastly, the influence law of the changes in pipeline foundation and underlying soil layer on pipeline settlement caused by two excavation methods is discussed.

2. Project Overview

2.1. Tianjin Metro Station

Xinxingcun Station of Tianjin Metro Line 4 is located in Xinxing Village of Dongli District. Figure 1 shows the plane layout of the foundation pit and pipeline. The station foundation pit is 465.22 m long, 19.7 m wide in standard sections and 36.75 m wide at maximum, and 13.7 m deep. On the north side of the foundation pit, there is a DN200 aviation oil pipeline. The minimum clear distance from the pipeline to the edge of the foundation pit is 5.81 m. The pipeline runs along the foundation pit at a 0.7–1 H depth range. It is buried about 1.53 m deep in the clay stratum, the soil below which is muddy soft soil sensitive to external disturbances.

2.2. Engineering Geology

The proposed site is underlain by Quaternary strata. The soils on the site include plain fill, clay, muddy soil, silty clay, and sandy clay. The soils in the upper 15 m of the strata are mainly clay, muddy soil, and soft silty clay, with silty clay being dominant. The thickness of each soil layer and the relief of their top and bottom are similar. The part in the lower 20 m features alternating sandy silt and silty clay, silty clay is in a plastic state, and sandy silt is dense with low compressibility. The relief of the top and bottom of each layer varies slightly.

In the upper 15 m thick sedimentary layer, the soils have high compressibility, with elastic modulus ranging from 8 MPa to 15 MPa. The void ratio is high, varying from 0.85 to 1.3. The plastic limit is in the 0.9–1.25 range. In the lower 20 m thick sedimentary layer, the soils have a high degree of consolidation and low compressibility, and the elastic modulus ranges from 17 MPa to 28 MPa. The void ratio and the plastic limit vary from 0.60 to 0.84 and from 0.4 to 0.8, respectively. Figure 2 shows the soil properties across the soil profile.

The aviation oil pipeline in this project is located in a soft clay layer with poor soil texture and high compressibility. Below the clay layer is muddy soil. The muddy soil layer mainly comprises muddy clay distributed stably on the first marine bed. With high sensitivity and low strength, this soil layer is prone to creep and disturbance. Engineering geological conditions are poor. Therefore, the excavation of deep foundation pits can easily disturb adjacent existing pipelines.

2.3. Supporting Method

The sides of the foundation pit are protected with Φ 1000 @ 1200 bored piles and 600 mm thick TRD waterproof curtain. C20 mesh shotcrete with a thickness of 15 cm is applied between the piles. The bored piles are topped with 1200 mm × 1100 mm reinforced concrete top beams. A total of three supports including one concrete support and two steel supports are set up on the basis of the excavation depth. The concrete support is provided with 1400 mm × 1200 mm concrete middle beams. The first group of reinforced concrete supports have a section size of 0.8 m × 0.8 m and are spaced at an interval of about 7 m. The second and third groups of steel pipe supports and steel supports (for change purposes) have a size of Φ 800 × 16; they are spaced about 3 m apart and subjected to 50 kN and 100 kN prestress, respectively. A schematic diagram of the support structure is shown in Figure 3.

3. Numerical Simulation

3.1. Excavation Method

Different excavation methods have different influences on the surrounding soil environment of the foundation pit. The aviation oil pipeline is laid on soft clay. This soil layer and the lower layer of muddy clay show poor texture, high compressibility, and high sensitivity. These factors can contribute to a greater impact on the pipeline, depending on the excavation method. Research on the influence of different excavation methods on soil in the soft soil area can lead to more effective construction schemes. This project is located in an urban center, imposing significant restrictions on the excavation of the foundation pit. Given the shape and surroundings of the foundation pit, this paper presents two different excavation methods: layered excavation and bench excavation (see Figure 4). In layered excavation, the soils to be excavated for the foundation pit are divided into four layers, each of which consists of 14 sections. The excavation is performed in sections from two ends to the middle. In bench excavation, the excavation is carried out in four tiers synchronously in a direction away from the pipeline. Each tier is divided into 14 sections. In both excavation methods, supports are provided promptly following the completion of the excavation of each section. The soil at each layer is excavated to 0.5 m below the support structure of that layer. The excavation depth of each layer is 1.5 m, 4.5 m, 9.2 m, and 13.7 m.

3.2. Numerical Model

The finite element method based on the continuous medium model was adopted with the following assumptions:

(1)

The influence of time on soils is not considered during the excavation of foundation pit.

(2)

When the station foundation pit is excavated, the foundation pit dewatering has been completed. Water permeability is not considered separately.

(3)

The pressure of the liquid inside the aviation oil pipeline on the pipeline body is directly replaced by area load.

(4)

Contact surface parameters are not set between materials at different strata. There is no relative slip between materials at different strata, and their displacement is coordinated.

(5)

The initial stress field covers gravity stress.

The soils in this project were considered continuous, uniform, and isotropic. The Mohr–Coulomb failure criterion was adopted for modeling. The elements were of solid type. According to the geological exploration period data, the physical properties of the soil are shown in

Table 1. Physical properties of soil.

Soil Layer	Average Thickness (m)	γ (kN/m3)	φ (°)	c (kPa)	$\mathsf{\mu }$	ω (%)	Void Ratio e	E (MPa)
Clay	2.1	18.1	8.5	19.4	0.3	42.1	1.241	12
Muddy clay	3.5	17.5	6.7	7.6	0.35	45.1	1.309	8.5
Silty clay	7.0	19.0	12.9	12	0.35	30.4	0.873	15
Clay	1.9	18.0	9	14.1	0.33	29.2	0.862	15
Silty clay	1.7	20.1	12.3	14.2	0.35	23.1	0.673	17
Silty clay	1.9	20.4	12.1	13.8	0.35	22.3	0.674	16
Sandy silt	4.7	20.1	29.9	15.9	0.3	21.7	0.642	23
Silty clay	5.9	19.2	10.8	24.6	0.32	28.8	0.835	20
Silty clay	3.0	19.8	13	15	0.34	23.2	0.694	22
Sandy silt	1.6	20.3	29	14	0.31	19.7	0.607	28

. As the section area of steel reinforcement in bored piles of the support structure was small, the elastic modulus of the poured concrete was taken as the elastic modulus of the pile body. For the convenience of calculations, the bored pile in the foundation pit was equivalent to the wall structure of the same material, and the equivalence principle was equivalent to the inertia distance of the section. The equivalence equation is shown below.

$$\frac{1}{64}\pi D^2=\frac{1}{12}B{h}^3,$$

(1)

where h is the thickness of the equivalent wall structure (m), D is the diameter of a bored pile (m), and B is the spacing of the originally designed bored piles (m). The equivalent thickness of the bored piles was calculated to be 0.8 m. The properties of each element in the model are shown in

Table 2. Physical properties of elements.

Material	Element Type	Constitutive Relation	E (GPa)	$\mathsf{\mu }$	γ (kN/m3)
Waterproof curtain	Plate element	Elasticity	0.015	0.20	21.5
Bored pile	Plate element	Elasticity	31.5	0.20	25
DN200 pipe	Plate element	Elasticity	210	0.30	78.5
Concrete support	Beam element	Linear elasticity	31.5	0.20	25
Steel support	Beam element	Linear elasticity	210	0.30	78.5

; the physical parameters of each component were determined empirically according to the material characteristics of the component.

According to Saint–Venant’s principle and engineering experience [24], the sphere of influence of foundation pit excavation was 3–5 times that of the excavation depth. The model extended no less than 50 m in all directions according to the dimensions of the station foundation pit. The overall size of the model was 580 m × 140 m × 70 m (x × y × z). The model adopted a hybrid modeling approach with hexahedrons and tetrahedrons. The calculation model is shown in Figure 5, where the black wireframe represents the soil element, and the colored elements model the foundation pit. An interface element was set at the outside of the pipeline body element, with the aim of maintaining non-coordinated deformation between the pipeline and the soil, without the constraints established by the model setup. The side and bottom boundaries of the model were fixed both horizontally and vertically, and the top surface had a free boundary. The simulation of foundation pit excavation was realized through the “grid passivation” function in Midas; that is, when a certain part of soil was excavated, the soil element was removed to simulate the excavation of soil.

3.3. Result Analysis

3.3.1. Deformation of Adjacent Soil

Prior to the simulation of the construction stage, the initial state of the model was corrected for zero displacement. Moreover, the initial state stress was analyzed, with the initial stress field used for the initial state. The paper, which focuses on a pipeline adjacent to a foundation pit, mainly deals with soil deformation outside the support structure, without researching the soil deformation inside the foundation pit. Figure 6 shows the horizontal displacement induced by excavation. It can be seen from the figure that, in both layered excavation and bench excavation, the maximum horizontal displacement occurred mainly in the middle part of the foundation pit. For ease of analysis, three points were selected in the figure. Point 1 is the midpoint of the area where foundation pit excavation begins. Point 2 is the midpoint of the area near the side of the pipeline. Point 3 is the midpoint of the tapered section. All three points are located at the edge of the foundation pit.

According to the numerical simulation results of layered excavation, the maximum horizontal displacement of soil was 2.7 mm when the foundation pit was excavated to a depth of 1.5 m, and the maximum point was 0.7 m below the surface; the excavation depth was 4.5 m, the maximum horizontal displacement was 4.8 mm, and the maximum point was 3.2 m away from the ground; the excavation depth reached 9.2 m, the maximum horizontal displacement was 7.6 mm, and the maximum horizontal displacement point was 11.4 m away from the ground. Figure 6a shows the horizontal displacement of the soil after the excavation was completed (at a depth of 13.7 m). The maximum horizontal displacement was 11.7 mm and occurred at 15.2 m below the ground at point 2. It can be seen that, with the progress of excavation, the position of the peak point of horizontal displacement continued to decline, with the depth of decline being 46.7%, 106.7%, 140.4%, and 131.1% of the excavation depth increment. It can be found that a greater excavation depth increment led to a greater decline in the peak point of horizontal displacement. A smaller thickness of the excavation layer led to a lesser fall of the horizontal displacement point, enabling the support structure to play its full part. However, a larger depth of excavation facilitates construction. The thickness of an excavation layer should be designed on the basis of the requirements of the project. The simulation results of bench excavation show that, when the excavation progress reached half of the total workload, the maximum horizontal displacement reached 5.14 mm and occurred at point 2, 13 m below the ground. The horizontal displacement after the excavation was completed is shown in Figure 6b. The maximum horizontal displacement was 11.1 mm and occurred at a point along the width of the tapered portion (i.e., point 2 in the figure). This point with peak displacement was 15.5 m below the ground.

Figure 7 shows the horizontal displacement at the three points. From the figure, it can be concluded that the displacement at point 1 was smaller than that at the other two. Li Dapeng et al. [28] analyzed the spatial effect mechanism and soil stress. They believed that the soil arching effect in the areas near the corners of a pit reduced soil pressure, which contributes to deformation reduction for the support structure and soil. For a long, narrow deep foundation pit in metro transit projects, excavation at the ends of the pit sees small deformation, despite great excavation depths. The deformation control by the support structure focuses on the lateral displacement control of the continuous wall in standard sections in the middle of the pit. At point 1, bench excavation caused greater horizontal displacement than layered excavation, and the horizontal displacement in both excavation methods showed basically the same change trend. Analysis of the displacement at different depths at each location shows that the maximum displacement at points 2 and 3 was about 15 m below the surface, the displacement at the surface was the minimum, and the displacement of step excavation was greater than that of layered excavation. The maximum displacements caused by step excavation at points 2 and 3 were 11.1 mm and 10.4 mm respectively, while the maximum displacements caused by layer excavation at points 2 and 3 were 10.6 mm and 9.3 mm, respectively. The horizontal displacements at points 2 and 3 varied significantly along the depth, varying by 8.6 mm and 9.6 mm for the layered excavation method and step excavation method at point 2, and by 5.2 mm and 7.1 mm at point 3, respectively. This indicates that the layered excavation method has little influence on surrounding soil and is preferable.

3.3.2. Pipeline Settlement Analysis

The reaction of the aviation oil pipeline adjacent to the foundation pit for soil excavation was analyzed. A total of 20 points were set along the pipeline in the x-axis direction (see Figure 3) at intervals of 20 m, with the end O of the foundation pit as the starting point. Figure 8 shows the variation pattern of pipeline settlement in the four stages during excavation. It can be found that the distance between the foundation pit and the pipeline had a significant impact on pipeline settlement. The maximum settlement of the pipeline occurred in the section adjacent to the foundation pit. Upon completion of the excavation, layered excavation resulted in pipeline settlement of 3.2 mm at minimum and 8.1 mm at maximum. The final pipeline settlement in bench excavation showed a similar pattern to that in layered excavation, with a large settlement in the middle and a small one at both ends. Bench excavation caused a pipeline settlement of 3.4 mm at minimum and 12.5 mm at maximum, with the amount of variation reaching 9.1 mm. The overall pipeline settlement caused by bench excavation was larger than that caused by layered excavation. The settlement of pipeline caused by bench excavation showed great variation, which is not conducive to the safety of the pipeline.

Point A, 70 m away from point O, was the pipeline point nearest to the foundation pit. Figure 9 shows the changes in vertical pipeline settlement at point A. The curve presenting the variation of pipeline settlement with the construction stage of layered excavation was in the form of steps. The vertical displacement of the pipeline experienced several abrupt changes, from 14.41 mm at the initial stage to 5.25 mm at the final stage. These abrupt changes were caused by the advancement of the excavation face of foundation pit soil toward the cross-section at point A. The settlement changes caused by each excavation to point A were 0.73 mm, 2.68 mm, 2.62 mm, and 2.11 mm. In bench excavation, the pipeline settlement curve showed a single abrupt change. At the third excavation stage, when the excavation face reached the cross-section at point A of the aviation oil pipeline, the pipeline settlement started to rise from 14.20 mm. At the seventh stage, the excavation face passed through the cross-section at point A, and the vertical pipeline settlement fell to a minimum of 6.09 mm. At the eighth stage, the vertical pipeline displacement was 7.60 mm due to soil creep. In the subsequent stages, the vertical displacement remained at 7.60 mm. The maximum amount of variation throughout the excavation stages was 8.11 mm. By comparing the two excavation methods, it was found that layered excavation led to small vertical pipeline displacement in both magnitude and rate of change; therefore, it is more preferable for pipeline safety.

According to the analysis of horizontal displacement and settlement of soil near the pipeline, the maximum horizontal displacement of the soil near the foundation pit caused by layered excavation was only 0.5 mm less than that caused by step excavation, while the maximum settlement of the soil near the foundation pit caused by layered excavation was 35.2% (4.4 mm) less than that caused by step excavation. The above shows that layered excavation had slightly better control over horizontal displacement of soil than step excavation, while layered excavation had much better control over settlement than bench excavation. This may be because the step excavation was carried out in the same direction, and the depth of the foundation pit excavated first was larger, while the depth of the excavation later was smaller, resulting in greater settlement of the soil near the larger depth. Layered excavation always maintains a uniform depth along the length of the foundation pit; thus, this excavation method has a great difference in the influence of the nearby soil displacement. Overall, the layered excavation method is preferable with obvious advantages in terms of pipeline settlement, soil disturbance, and safety of the support structure. In the actual construction of this project, it was more advantageous to choose the layered excavation method; Figure 10 shows the panorama of actual engineering excavation.

3.3.3. Analysis of Pipeline-Soil Interaction

Excavation of foundation pits can cause the displacement of a pipeline, soil–pipeline detachment, and pipeline suspension. These problems pose considerable risk to pipeline safety. Since the existing pipeline is buried underground, it is difficult to measure the pipeline-soil gap directly. According to the layered excavation model, this paper analyzes whether measures should be taken for the pipeline to predict the pipeline-soil gap by calculating the gap size.

Figure 11 shows the state of detachment between the pipeline and soil. According to the numerical analysis results, the deformation of the pipeline itself was small; its displacement was mainly driven by the deformation of the surrounding soil, resulting in gaps between the pipeline and soil. For ease of exposition, this paper defined the vertical upward direction as 0° direction, and the angle increased clockwise. From the figure, it can be concluded that, at the beginning of the excavation, the pipeline and soil had a large gap at 30° and 210°, and the pipeline and soil fit closely at 120° and 300°. As the foundation pit excavation depth increased, the pipeline-soil gap was closed at 30° and 210°, while a large gap appeared at 120° and 300°.

In order to clearly explain the influence of gap size between the pipeline and soil on the pipeline safety performance, it is necessary to quantitatively analyze the variation pattern of pipeline gap width. However, the numerical model only enabled us to determine the vertical and horizontal displacements and total displacement of each point of the pipeline, and the pipeline-soil gap width could not be obtained directly through the model. To obtain the gap size between the pipeline and soil, the horizontal and vertical displacements were treated as follows:

$$d=L\cos \left(\alpha +\theta -\frac{\pi }{2}\right),$$

(2)

where d is gap width (mm), L is the detachment distance between soil and pipeline (mm), $L=\sqrt{{\left(y^{\prime }-y\right)}^2+{\left(x^{\prime }-x\right)}^2}$, and θ is the angle (°) of the pipeline, with the vertical upward direction defined as 0°. The angle increased clockwise. Furthermore, α is the displacement angle (°) of a point, $\alpha =\arctan \left(\frac{y^{\prime }-y}{x^{\prime }-x}\right)$, in which x represents horizontal displacement and y represents vertical displacement (mm).

As shown in Figure 12, substituting L (i.e., detachment distance between soil A′ and pipeline point A) and α into Equation (2), the gap width between the pipeline and soil could be calculated (see Figure 13). In Figure 13, the value with a dotted line indicates the thickness of the soil squeezed between the pipeline and soil. A smaller dotted line value denotes a greater squeeze. As can be seen from the figure, the peak values of the gaps in the four stages of the excavation process were 1.6 mm, 1.1 mm, 1.2 mm, and 2.3 mm, occurring at 25°, 65°, 108°, and 110°, respectively. It can be found that, as the excavation depth increased, the peak values of the gaps first decreased and then increased, and the angle of the positions with peak values changed from 25° to 110°. In the whole excavation process, the pipeline and soil exhibited certain relationships. Because of the influence of gravity and soil disturbance, the pipeline-soil gap changed from an open state to a squeeze state at the angles of 25° and 205° to the pipeline. When excavation was completed, the soil was squeezed to the maximum. At the angles of 110° and 205° perpendicular to the gravity direction, the pipeline-soil gap changed from a squeeze state to an open state. The gap width reached the maximum value after the foundation pit excavation was completed.

Made of steel, the pipeline has stiffness strong enough to resist deformation due to stress distribution changes caused by pipeline suspension in the whole process of foundation pit excavation. The stress distribution changes have little impact on the safety and stability of the pipeline. Upon completion of foundation pit excavation, the soil around the pipeline should be changed to ensure the pipeline continues to operate safely and stably.

4. Field Test and Validation

4.1. Test Scheme

The field observation aimed to measure the displacement of the foundation pit soil and pipeline. Since the direct arrangement of measuring points on the aviation oil pipeline can disturb the pipeline greatly, 23 points (GXC-1 to GXC-23) for measuring pipeline settlement were set on the soil above the aviation oil pipeline, and five points (TST-1 to TST-5) for measuring horizontal displacement were set within the nearest range of the station pipeline and foundation pit, with the specific location of each point shown in Figure 14.

In this paper, measuring point refers to displacement measuring point mainly used for the measurement of surface vertical displacement and stratum horizontal displacement. Surface vertical displacement is mainly measured with a level gauge. Before the construction of support structure, three measurements were carried out by vertical displacement measuring points already arranged in the surrounding environment, and the mean value was taken as the initial value within the tolerance range. In the event of measurement of stratum horizontal displacement, an inclinometer was used to measure the deformation of the inclinometer pipe buried in the soil layer from bottom to top in order to make clear the horizontal displacement of soil layer around the envelope structure at different depths during the excavation of the foundation pit, as shown in Figure 15. During measurement, the inclinometer probe went down slowly to the bottom of hole along the guide groove, so as to measure the horizontal displacement every 500 mm from bottom to top. To figure out the deep lateral deformation, the fixed starting point, supposedly located at the top of the inclinometer pipe, was determined. Pipe orifice displacement was measured by a total station; for calculation, the actual displacement of orifice was corrected.

4.2. Horizontal Displacement of Soil Adjacent to Pipeline

The unloading of soil in the foundation pit will also have an effect on the pipeline. Since the horizontal displacement monitoring point could not be directly placed on the pipeline, the influence of soil excavation in the foundation pit on the pipeline horizontal displacement was studied by analyzing the horizontal displacement of soil mass around the pipeline. Therefore, five horizontal displacement monitoring points were set in the soil near the foundation pit and pipeline, as shown in TST-1 to TST-5 in Figure 16.

Figure 16 shows a comparison of the actual horizontal displacement and the simulated displacement of soil at the measuring points of TST-1, TST-3, and TST-5. It can be seen from the figure that the numerical simulation results kept a trend of consistency with the actual measured results, and the overall change law of horizontal displacement at the three measuring points remained similar. With the increase in excavation depth, the horizontal displacement peak of each measuring point also increased. However, the position of the horizontal displacement peak point at each measuring point declined; at the beginning of excavation, the horizontal displacement peak point was located 2.5–3.0 m below the surface, and the horizontal displacement peak point dropped to the maximum depth of the measurement range (6 m below the surface) after excavation was completed. In practice engineering, the reasonable arrangement of internal support near the maximum point of horizontal displacement can help better control the deformation of soil, give full play to the self-confining effect of foundation pit excavation, and mitigate the disturbance to soil, thereby indirectly achieving the effect of pipeline protection. For TST-1, TST-3, and TST-5, their actual displacement peak increased from 1.85 mm to 8.19 mm, from 2.17 mm to 14.09 mm, and from 2.31 mm to 11.23 mm, with increases of 6.37 mm, 11.92 mm and 8.92 mm, respectively, and their distance to the edge of the foundation pit was 9.21 m, 2.35 m, and 5.13 m, respectively. This means that as the measuring point approached the foundation pit, the horizontal displacement peak became greater. According to the actual situation, the minimum clear distance between the pipeline and the edge of the foundation pit was 5.81 m, and the distance of between the buried site and the surface was 1.53 m. Therefore, on the basis of the analysis of the horizontal displacement of each measuring point, it can be speculated that the actual maximum horizontal displacement of the pipeline was lower than 6.5 mm, which is within the safety control limit of pipeline horizontal displacement.

4.3. Soil Settlement at the Top of Pipeline

As shown in Figure 17, the settlement caused by the construction of the support structure was taken as the initial settlement, and the numerical analysis results were compared with the actual measured results. Since it was impossible to consider all factors in the numerical analysis, the actual measured results were different from, i.e., lower than, the numerical results to some extent. In terms of the variation trend, the actual measured results were similar to the numerical analysis results, indicating that the numerical analysis results were of high reliability. It can be seen from the figure that, with the advance of the construction stage, the vertical settlement of GXC-03, GXC-08, and GXC-10 measuring points gradually decreases, the settlement value of the GXC-03 point tended to be stable, and the settlement value of the GXC-16 or GXC-21 measuring point showed an upward trend, mainly attributable to the difference in distance between the pipeline and foundation pit. Hsieh et al. [29] divided the settlement area of surrounding soil caused by foundation pit excavation into two parts, i.e., the major influence area and secondary influence area, with a critical value y₀/H_e in the middle (y is the length from foundation pit support, and H_e is the depth of foundation pit). A lower y/H_e than the critical value indicates a large soil settlement and obvious change in settlement; a higher y/H_e than the critical value represents a small soil settlement and slow change in settlement. Figure 18 shows the settlement cross-section at the center of the model. The soil 15 m outside the foundation pit was the major influence area, where the settlement of the earth surface gradually decreased with the increase in excavation depth. The soil beyond 15 m was the secondary influence area, where the surface settlement increased with the increase in excavation depth.

For GXC-08, the nearest measuring point to the edge of foundation pit, the maximum settlement of soil at the top of the pipeline was 14.14 mm according to numerical simulation, the change in settlement was 7.55 mm during excavation, and the settlement arising from excavation at each layer accounted for 13.2%, 31.3%, 27.7%, and 27.8% of the total settlement. In actual excavation, the maximum settlement of soil at the top of the pipeline was 14.27 mm, and the change in actual settlement was 5.81 mm, and the change in settlement for each soil layer was 2.01 mm, 2.42 mm, 0.59 mm, and 0.79 mm, accounting for 34.6%, 41.6%, 10.2%, and 13.6% of the total settlement, respectively. The settlement variation range caused by the excavation of the first and second soil layers accounted for three-fourths of the total settlement, revealing that the initial excavation of soil caused a great disturbance to the adjacent area. For GXC-21, a measuring point far away from the foundation pit, the maximum settlement of soil at the top of the pipeline was 5.70 mm according to the actual measurement, and the overall settlement variation range was 2.12 mm. The maximum settlement was 4.80 mm according to simulation, and the overall settlement variation range was 1.56 mm during excavation. It can be seen that the distance between the pipeline and foundation pit was negatively correlated with the settlement of soil at the top of the pipeline.

By comparing the actual measured data and numerical analysis results of horizontal displacement and settlement at each monitoring point, it can be found that, due to some assumptions made in the numerical modeling, the variation trend of the numerical simulation results was relatively simple, while the field measured data fluctuated greatly with the foundation pit excavation, such as at the measuring points TST-1, GXC-10, and GXC-13. On the whole, the variation trend of the numerical simulation results was basically consistent with the field measurement data, and the numerical simulation results could reflect the soil displacement variation law in the actual excavation process. In terms of data accuracy, there was a significant deviation between the numerical simulation results of some monitoring points and the onsite measurement data, but the deviation values were within an acceptable range, and the accuracy of most points was relatively high. The above results indicate that the reliability of the numerical analysis results was high, and the model established in this paper was reasonable and reliable; thus, it can be used as the basis for analyzing the influence of foundation pit excavation on adjacent pipelines.

5. Discussions

The preceding analysis indicated that layered excavation led to evident advantages in pipeline settlement and soil disturbance support structure safety, while full-section bench excavation caused a great disturbance of the surroundings. Therefore, it can be concluded that the bench excavation method was inferior to the layered excavation method in this project. The performance of soil around the pipeline also played a key role in helping the pipeline resist soil disturbance. The buried depth of transmission pipeline was generally not more than 3 m, and the soil around pipeline was usually the soil layer close to the surface. To further understand the influence of foundation pit excavation on aviation oil pipeline in different soils, the physical properties of three common soil layers near the surface, i.e., plain fill layer, clay layer, and silty soil layer, are set out in

Table 3. Physical properties of different soil layers.

No.	Soil Texture	Mass (kN/m3)	Angle of Internal Friction (°)	Cohesion (kPa)	Elastic Modulus (MPa)	Poisson’s Ratio
Soil layer 1	Plain fill	18.1	8.5	19.4	6.0	0.2
Soil layer 2	Clay	18.1	8.5	19.4	12.0	0.2
Soil layer 3	Silty clay	18.1	8.5	19.4	12.0	0.35

. The displacement rules of the pipeline caused by different soil layers and excavation methods were analyzed by changing the properties of the pipeline soil layer and its underlying soil layer in Figure 2.

According to the coordinate system given in Figure 3, a total of 20 points were set along the pipeline in x-axis direction at intervals of 20 m. Through numerical calculation based on the physical properties in Table 3, the comparison of pipeline settlement in different soil layers is shown in Figure 19.

Figure 19 shows that layered excavation generated a smaller pipeline vertical displacement than bench excavation under different soil layer properties, with apparent advantages in terms of control of displacement variation. The vertical displacement of the pipeline caused by the two construction methods tended to be larger in the middle and smaller at both ends with the location of the measurement point, which was mainly related to the excavation shape of foundation pit and the distance between pipeline and foundation pit. With the elongated shape of the station foundation pit of the subway station, the excavation from one end could only cause soil disturbance to the middle section in the process of excavation, further resulting in the soil settlement superposition phenomenon in the middle section, while finally presenting the settlement trend of large in the middle and small at both ends. Because of the different distances between the foundation pit and the pipeline, the pipeline behind the No. 17 measuring point had a relatively high distance from the foundation pit; the secondary influence area was less affected by foundation pit excavation and confronted with normal soil settlement, leading to the secondary decline of the curve behind the No. 17 measuring point.

To more intuitively compare the pipeline settlement caused by the two excavation methods under different soil layer properties, the maximum, minimum, and average values of pipeline settlement under soil layer properties are summarized in

Table 4. Pipeline settlement of different soil layers.

Construction Method	Name of Soil Property	Plain Fill	Clay	Silty Clay
Layered excavation	Max. (mm)	6.62	7.16	8.97
	Min. (mm)	3.34	3.19	2.21
	Avg. (mm)	5.37	5.53	6.14
Bench excavation	Max. (mm)	8.81	9.87	12.46
	Min. (mm)	4.07	3.51	3.36
	Avg. (mm)	6.73	7.05	8.23

.

Table 4 shows that, compared with the bench excavation method, the layered excavation method had lower maximum, minimum, and average settlement values of the surface median line under different soil layer properties. Furthermore, the layered excavation method also enjoyed a smaller range of settlement variation than the bench excavation method. In general, the layered excavation method was superior to the bench excavation method in terms of pipeline settlement control.

Figure 20 compares the pipeline settlements caused by the same excavation method in different soil layers. According to a comparison of the properties of different soil layer properties in the same excavation method, the pipeline settlement values in plain filled soil and clay were close, while the settlement values in silty clay were larger than those in the other two soil layers. According to a comparison of the physical properties of plain fill and clay, the main property difference is the elastic modulus, with that of plain fill being half that of clay. A lower elastic modulus improves the compressibility of soil and increases the settlement of pipeline. However, in general, the settlement values of plain filled soil and clay were similar, and the elastic modulus exerted little influence on pipeline settlement. By comparing the physical properties of clay and silty clay, under the same elastic modulus, different Poisson’s ratios resulted in a great difference in pipeline settlement, as shown in the figure. A larger Poisson’s ratio of soil indicates a greater deformation capacity and increased sensitivity to excavation disturbances of surrounding soil, thus resulting in greater pipeline settlement in silty clay than in plain fill and clay.

In summary, according to the analysis and discussion of the settlement deformation of different pipeline soil layers and its underlying soil layers under different excavation methods, suggestions are given for adopting different construction methods under different construction conditions. When excavating the foundation pit near an existing pipeline, if the elastic modulus of the soil layer near the pipeline and the underlying soil layer is small and Poisson’s ratio is large (i.e., the soil layer has high compressibility and deformation), the layered excavation method should be used for construction. If the elastic modulus of the soil layer near the pipeline and the underlying soil layer is large and Poisson’s ratio is small (i.e., the soil layer has low compressibility and deformation), the impact of the two excavation methods on the pipeline is close. In this case, the emphasis of the station construction requirements should be weighed, and the appropriate construction method should be selected on the basis of the construction period, project cost, and other factors. Compared with the characteristics of layered excavation and bench excavation, the bench excavation method has the characteristics of large excavation surface, convenient construction, and fast speed, which is conducive to actual projects. Therefore, it is recommended that the bench excavation method be used for the excavation of soil layers with low compressibility and deformation.

6. Conclusions

Considering the soft soil feature of the aviation oil pipeline adjacent to the Xinxingcun Station of Tianjin Metro Line 4 in China, this paper compared and analyzed the differences in the impact of layered excavation and step excavation on the adjacent pipeline through Midas numerical simulation, and then discussed the impacts of foundation excavation on the pipeline-soil gap, before comparing the numerical simulation results with the measured data. Lastly, this paper discussed the influence of changes in the pipeline foundation and underlying soil layer on the settlement of the pipeline. The main conclusions are drawn as follows:

(1) The layered excavation method has better control over horizontal displacement and settlement of pipelines than the bench excavation method. During the excavation process, the maximum vertical settlement of the aviation oil pipeline caused by layered excavation and bench excavation was 8.1 mm and 12.5 mm, and the latter was 54.3% more than the former. The settlement of pipeline tended to be large in the middle and small at the ends, regardless of the bench excavation method or layered excavation method.

(2) Excavation of the foundation pit can cause the pipeline near the foundation pit to partially separate from the surrounding soil. Cracks occurred in the vertical direction of the pipeline after the construction of the foundation pit support structure. In the process of foundation pit excavation, the pipeline-soil gap changed from an open state to a squeeze state at the angles of 25° and 205° to the pipeline. When excavation was completed, the soil was squeezed to the maximum at the angles of 110° and 205° to the pipeline, the pipeline-soil gap changes from a squeeze state to an open state, and the pipeline-soil gap reached its maximum value of 2.3 mm.

(3) With the advance of the construction stage, the vertical settlement of GXC-03, GXC-08, and GXC-10 measuring points gradually decreased, the settlement value of GXC-13 point tended to be stable, and the settlement value of the GXC-16 and GX-21 measuring points showed an upward trend. The comparison of the monitoring data showed that the numerical simulation results were basically consistent with the field measurement data. Therefore, it is proper and feasible to use the layered excavation method for the foundation pit excavation of Xinxingcun Station.

(4) The comparison of different soil layer properties showed that layered excavation had a better effect on controlling pipeline settlement than bench excavation. In the event that the soil layer where the pipeline is located is of high compressibility and deformation, the layered excavation method is the first choice. Otherwise, it is necessary to grasp the key points in station construction and comprehensively select an appropriate construction method.