Abstract

Natural gas pipeline projects in mountainous areas are inevitably affected by geological disasters such as landslides, which pose a serious threat to the safe operation of pipelines along the routes crossing landslide areas. In this paper, based on a pipe-landslide project in a mountainous area in southwest China, the interaction mechanism and failure evolution process of the landslide-pipeline system reinforced by two kinds of micropiles are studied through indoor large-scale physical model tests, and some suggestions on the support work of the pipe-landslide project are put forward according to the test results. It was found that the deformation process of the engineering system composed of landslide, micropile, and pipeline presents a high degree of synergy under the external force and mainly experiences four stages: initial deformation period, uniform deformation period, accelerated deformation period, and residual deformation period. The bending deformation of the perforated pipe micropile is large at the 1/4 position of the pile top from the pile bottom, and the deformation of the screw micropile near the sliding surface is serious. The pipeline welding port is the weak position of the pipeline; after the failure of the pile, the pipeline interface is first cracked, along the interface position along the two ends of the tear, and finally completely broken. The screw micropile cannot effectively resist the landslide thrust at a large load level, so the risk of pipeline damage is greater. The yield strength and ultimate strength of the perforated pipe micropile are greater than those of the screw micropile, and the perforated pipe micropile can still exert a certain residual resistance after reaching the ultimate bearing capacity, which has a beneficial effect on the reinforcement of the pipeline crossing the landslide system. The research results provide important reference value for landslide-pipeline treatment engineering.

1. Introduction

Since the beginning of the 21st century, with the vigorous development of the global economy, the energy demand of the world is increasing. Natural gas, as a clean and environmentally friendly high-quality energy, has been valued and favored by governments of various countries. Natural gas pipeline projects in many countries are developing and constructing rapidly. The pipeline transportation of natural gas has ensured the daily needs of many residents.

The geological conditions along the natural gas pipeline project in mountainous areas are complex. It is difficult to avoid crossing the unfavorable geological areas during the construction process, and it is very likely to cause geological disasters under the influence of unstable factors. Landslide is one of the common geological disasters in the oil and gas pipeline project in mountainous areas [1]. The impact of landslide on pipeline engineering cannot be underestimated. Under the action of landslide thrust, the pipelines buried in the slope are prone to deformation or even damage, resulting in gas leakage, which seriously threatens the safety of pipelines along the line and affects the normal life of the people. According to the relevant departments of the United States from 1991 to 2014 natural disasters caused by pipeline accidents, landslides accounted for about 54% of natural disasters, accounting for 1.8% of the total number of pipeline failure [2]. Therefore, effective control measures should be taken to protect the pipeline safety when the pipeline passes through the landslide area. Topal and Akin [3] studied the stability of pipeline landslide in Karacabey area by means of field investigation and measurement, laboratory test, and landslide limit equilibrium analysis and put forward the treatment methods such as removing landslide soil, leveling slope, and reducing buried depth of pipeline. Li et al. [4] studied the interaction mechanism between antislide piles and landslide soil and optimized the deployment of antislide piles to improve landslide stability and reduce pipeline stress and displacement. Marinos et al. [5] assessed the pipeline risk at 82 landslide sites in Albania, classifying landslides as “high risk,” “medium risk,” and “low risk” based on the assessment results and considering risk mitigation measures.

At present, there are still some problems in the treatment measures of pipeline-landslide system, such as difficult construction, serious disturbance to slope, and high cost. With the advantages of small pile diameter, flexible pile layout, simple construction, and good adaptability to various soil layers, micropiles have been widely used in landslide control projects. Scholars have carried out relevant research work on the design method and antisliding performance of micropile-reinforced slopes. Al-Obaidi and Al-Karawi [6] studied the influence of micropile length, diameter, grouting length, and soft soil type on load transfer and ultimate bearing capacity of micropile. Based on the assumption of sliding surface stress, Dong-ping et al. [7] established a new limit equilibrium method (LEM) to analyze the stability of antislide micropile-reinforced slope. According to the upper bound theorem of kinematic limit analysis, Zeng and Xiao [8] proposed an analytical method to calculate the net thrust of micropile group and the sliding surface of pile slope under the specified safety factor. In addition, Chao et al., Bian, and Sun et al. [911] have consistently shown that micropile has a good effect on reinforcing slopes.

In recent years, some new micropiles have been used in landslide treatment projects. There are mainly two new types of micropiles, as shown in Figure 1. One is the perforated pipe micropile, which is a new type of pressure grouting antislide pile. By applying pressure, the cement slurry is injected into the soil through the small holes on the steel pipe. When the slurry is solidified, it has cohesion and water absorption. The sliding body and the immobile body are bonded into a diffused complex to improve the geology. At present, scholars have less research on the antisliding performance of the perforated pipe micropile-reinforced landslide. Feng [12] proposed a theoretical calculation formula for landslide reinforcement by perforated pipe micropile. Wang and Shang [13] studied the double grouting technology of the perforated pipe micropile and proved that this technology can effectively improve the bearing capacity of micropile through experiments. Qiang et al. and He et al. [14, 15] have shown that the perforated pipe micropile has a beneficial effect on controlling the displacement of landslide soil, and high pressure grouting can improve the horizontal ultimate bearing capacity of the perforated pipe micropile-reinforced landslide. The second type is the screw micropile, which comes from the screw pile used in the foundation engineering. Because of its use of the principle that “screws are stronger than nails” [16], the screw pile has better bearing performance. However, the screw pile is only used in the foundation engineering and has not been used as a retaining structure in the landslide control project. Scholars mostly focus on the vertical bearing performance, such as Krasiński [17] who studied the bearing mechanism of screw pile by numerical simulation method and verified the reliability of the method by field load test. Malik et al. [18] carried out a comparative study on the bearing capacity of screw pile and straight pile. The results show that the bearing capacity of screw pile with shaft diameter ratio of 2-4.1 is 2-12 times higher than that of straight pile with similar shaft diameter. Zhen et al. [19] studied the bearing characteristics of screw pile in sand foundation by model test.

In summary, engineers and scholars have already had some research foundations in the interaction mechanism of pipeline-landslide and the antisliding characteristics of micropile-reinforced landslide, but there are few studies on the synergistic deformation mechanism of pipeline-landslide and supporting structure. In view of this, based on the natural gas pipeline-landslide control project in a mountainous area of southwest China, this paper simulates the antisliding characteristics of two kinds of micropile groups and the deformation characteristics of pipelines under the action of landslide thrust through large-scale physical model tests and tests the pile top displacement, pile strain, soil pressure on both sides of the pile, natural gas pipeline strain, and soil pressure on both sides of the pipeline. The spatial distribution of pile bending moment and peak earth pressure on both sides of the pile is further analyzed, and the deformation characteristics of the two micropiles are compared. Combined with the deformation and failure of the pipeline, the applicability of the two micropiles to strengthen the pipeline-landslide system is explored. The test results are summarized to provide theoretical reference for landslide disaster management along the pipeline.

2. Pipeline-Landslide Area Project Overview

The landslide site is located on a slope of Zhongwan Village, Huangchen Town, Cheng County, Longnan City, Gansu Province. The plane shape of the landslide is like an arm chair. The highest elevation of the trailing edge of the landslide is 1510 m, and the lowest elevation of the leading edge is 1470 m, with a relative elevation difference of about 40 m. The landslide is located in the upper part of the slope of the low mountain, the terrain is high in the east and low in the west, and the longitudinal slope is generally 10~35°. The slope surface is folded and stepped, and the slope direction is roughly to the west.

There are several arc-shaped tensile cracks in the posterior part of the landslide and multilevel staggered platform. The left side of the landslide is bounded by the original natural gully, and the right side is bounded by the microgeomorphologic changes. The leading edge of the landslide is not completely cut out, and the characteristics are not obvious, but several longitudinal cracks can be seen, and there are characteristics of microswelling. In the middle of the landslide, there were two obvious faults on the original cement road, one of which was located above the pipeline with a height difference of about 1.0 m. A wrong slope was found in the pepper field of the original natural gully on the left front of the landslide. According to the comprehensive analysis of the surface deformation characteristics of the landslide, it can be seen that the landslide is a composite landslide with the characteristics of partial traction. The landslide in the creep deformation stage at present, under natural conditions, is in stable condition; under the working condition of heavy rain and earthquake, landslide is in unstable state; under the action of external force, deformation of landslide may be further intensified; and the overall buckling gives natural gas in the pipeline and leading edge, threatening the local people’s life and property damage. The geological characteristics of pipe-landslide area are shown in Figure 2.

3. Model Test Design

3.1. Similarity Relationship Design

Physical model test is one of the important means to study the deformation and failure process of landslide [20], which can truly simulate the deformation and failure characteristics of landslide and the stress failure process of pipeline and pile under the reinforcement of micropile, reflect the mechanical properties of landslide rock and soil mass and structures qualitatively or quantitatively, and reveal the bearing mechanism of landslide strengthened by micropile. In order to ensure the accuracy and reliability of model tests, theorem established by Buckingham in 1914 is usually used to carry out experimental similarity design.

It is often difficult for natural phenomena to completely satisfy all the similarity conditions in the similarity law, so the approximate similarity method is usually adopted. According to the influence of various factors on the phenomenon in the experiment, the main factor is grasped and the secondary factor is omitted to carry out the experiment design [21].

Combined with the research goal of this paper, the stress and deformation characteristics of micropiles and pipelines under the action of horizontal thrust of landslide are studied. Therefore, the influence of gravity is neglected in the design test, and only the scale model is used. The geometric size of micropile, density, and gravity acceleration are the main control parameters, and the similarity ratios are 30 : 1, 1 : 1, and 1 : 1, respectively. The similarity ratios of other parameters can be derived according to the theorem, as shown in Table 1.

3.2. Proportioning Design of Similar Materials

The similar materials of landslide are more complex, and the similar relationships are not easily satisfied in the matching process. Therefore, when selecting the slide body ,slip belt, and bedrock materials, the basic physical quantity is based on several parameters that is the most likely to affect the stability of prototype landslide, such as bulk density, gamma, cohesion, internal friction angle phi c, and elastic modulus E. Referring to some research results obtained by other scholars [22, 23], through orthogonal ratio design and direct shear test and triaxial test to verify the accuracy and rationality of parameter values, it is determined that the mixture of laterite, quartz sand, cement, gypsum, and water is used to simulate grade IV bedrock, and the specific mass ratio is 70 : 30 : 5 : 3 : 10. The sliding body was stimulated by a mixture of laterite, quartz sand, and water with a specific mass ratio of 70 : 20 : 10. The sliding belt is simulated by a mixture of quartz sand, soil, talc, and water with a specific mass ratio of 27 : 52 : 35 : 15. The testing process of soil samples is shown in Figure 3(a), and the mechanical parameters of model materials obtained and their comparison with prototype materials are shown in Table 2.

In addition, it is difficult to make screw micropile and perforated pipe micropile with the same shape as the prototype. Therefore, the pile model is simplified, screw holes are drilled at equal intervals in the length direction of PP-R pipe, and pressure grouting cement slurry is used to simulate perforated pipe micropile. Similarly, the screw micropile is simulated using a plastic shell filled with cement paste. The cushion cap is simulated with hard wood. The natural gas pipeline is simulated by PVC pipe with a diameter of 20 cm, and the welding port of the pipeline is simulated by hot-melting the two sections of pipe to ensure that the model pipeline is as similar as possible to the prototype pipeline. The pile model and pipeline model are shown in Figure 3(b).

3.3. Model Test Equipment

According to the similar design, the size of the model box used in the test is (), and the box is composed of steel structure skeleton and tempered glass on both sides. The body is filled with a landslide model, which is divided into three layers from top to bottom: slide body, slide belt, and bedrock layer. Four groups of micropiles were embedded in the landslide body, with the center line of the box as the boundary. The two groups on the left were perforated pipe micropile, and the two groups on the right were threaded micropiles. A model of a natural gas pipeline is embedded in the slip body behind the pile. The rear side of the box body is fixed with a reaction wall, and a hydraulic jack is installed on the reaction wall. The jack exerts horizontal load on the landslide model by acting on the bearing plate. The model design is shown in Figure 4.

3.4. Sensor Layout and Model Filling Process
3.4.1. Sensor Layout

A complete real-time monitoring system is constructed according to the deformation characteristics of the structure and the evolution mechanism of landslide. The sensor mainly adopts high-precision dial gauge, resistance strain gauge, and earth pressure box. The monitoring data is collected by DH3816N static strain test system, which has 60 acquisition channels and can collect strain, earth pressure, and other test data at the same time.

In order to minimize the influence of the boundary effect of the model box, the micropile in the middle of the model was considered as the test object, and the deformation characteristics of the typical micropile were monitored by symmetrical resistance strain gauges at the same distance between the mountain side and the river side. Due to the great influence of temperature on resistive strain gauge sensor, a corresponding temperature compensation gauge is connected to each strain gauge. In order to test the mechanical properties of soil and rock mass on both sides of landslide reinforced by micropiles, soil pressure boxes are arranged symmetrically on the mountain side and river side of micropiles, and the sliding body, sliding surface, and bedrock layer are covered in depth. In addition, strain gauges were pasted symmetrically on both sides of the pipe model buried in front of the pile, and earth pressure boxes were arranged to test the stress and deformation of the pipe. The specific placement of sensors is shown in Figure 5.

3.4.2. Model Filling Process

The landslide model can be divided into three parts: bedrock, slide zone, and slide body. The embedded structures in landslide soil include perforated pipe micropile, threaded micropiles, and natural gas pipeline models. According to the spatial relationship of each rock and soil layer and structure, the elevation reference line is marked on the outside of the model box, and the filling of the test model is carried out from the bottom to the top, as shown in Figure 6. (1)Fill in the bedrock layer. Firstly, the bedrock model materials are mixed according to the above similar ratio relationship, and each basic material is weighed according to the ratio and placed in the mixer for mixing evenly and then filled at the bottom of the model box, and the designed pile burial position is reserved. In order to ensure that the model filling is consistent with the prototype, layered artificial compaction is carried out during filling, and soil samples are taken on site during the filling process to test the unit weight, internal friction angle, and cohesion to meet similar design requirements [24](2)Pile body is buried and fixed. The bottom of the trench reserved for the bedrock layer is leveled so that its height is consistent with the elevation of the bottom of the designed pile. After the single pile is fixed at the fixed position, it is put into the empty ditch for backfilling model soil. During backfilling, the soil around the pile bottom is compacted to keep it fixed. In the process of backfilling soil, soil pressure sensors are embedded in the soil pressure test positions on the front and back sides of the pile, and the surface of soil pressure is facing the loading side(3)Slide belt laid. After filling the bedrock layer to the design elevation, mix it according to the ratio of sliding zone model materials and evenly lay the prepared sliding zone soil on the top of the bedrock layer, smooth and compacted(4)Slip body filling. After the sliding belt is laid according to the design position, the sliding body model material is mixed and filled, and layered compaction method is also adopted during filling to ensure the accuracy of the test(5)Buried pipelines. After the sliding body is filled to the design depth of the pipeline, the pipeline model is placed horizontally in the design position, and earth pressure sensors are buried in the mountain and river sides of the pipeline, and finally, the soil is backfilled(6)Pile grouting. After the completion of landslide model piling, grouting of pile body is carried out. The screw pile side adopts self-made grouting funnel for gravity grouting. The cement slurry is pumped into the perforated pipe micropile model by pressure pump at the side of the perforated pipe micropile. After grouting is completed, the pile body is maintained in a static position, and the loading test is carried out after the strength reaches the requirements

3.5. Loading Condition Setting

The horizontal load was applied by hydraulic jacks in the test to simulate landslide thrust, and the loading method was hierarchical loading by slow maintenance loading method [25]. The initial loading pressure was 1 MPa, and the loading pressure at each stage was 0.5 MPa. A total of 7 loading conditions were set in the test. After each level of loading, the sublevel loading is carried out after the reading of the dial indicator on pile top is basically stable. The loading conditions are shown in Figure 7.

4. Analysis of Model Test Results

4.1. Macroscopic Deformation Characteristics of the Test Model
4.1.1. Description of Landslide Deformation Evolution Process

Most landslides have obvious deformation signs in the process of development and evolution. Due to various internal and external dynamics, surface cracks of different causes and types are formed in space. The occurrence sequence and spatial distribution of these cracks can well characterize the deformation process of landslides [26]. In order to more directly reflect the landslide evolution mechanism under the reinforcement of two kinds of micropiles, the apparent deformation characteristics of the model under different loading conditions were recorded and described. Typical experimental phenomena are shown in Figure 8.

Because the landslide soil itself has a certain ability to resist sliding, at the initial stage of loading, the thrust force exerted on the landslide is in balance with its own antisliding force, so that the landslide remains stable without obvious deformation.

With the increase of thrust force, the landslide deformation starts from the back edge of the slope body, and the back edge of the landslide first cracks, slips, and produces tensile cracks. On both sides of the back edge of landslide, due to the influence of boundary effect, shear stress and tensional stress are concentrated on the interface between soil and model box boundary, forming shear tension cracks, and the extension direction of cracks is 0°-20° from the horizontal direction of the side (see Figure 8(a)).

When the load on the back edge of landslide increases to 2.5 MPa, the deformation of rock and soil mass behind the pile develops continuously, the surface soil appears uplift, the overall number of cracks on both sides increases, the transverse length keeps increasing, and the width of cracks further expands. With the passage of time, a small number of cracks and uplift folds appear in the middle and lower part of the front slope body of the screw pile side and develop towards the perforated pipe micropile side (see Figure 8(b)).

When the pressure at the back edge of the landslide is increased to 3 MPa again, the stress of the soil behind the pile is concentrated due to the resistance of the antislide pile, resulting in extrusion deformation and swelling crack [27, 28]. The number of shear and tension cracks on both sides of the back edge of the landslide increases obviously, and they tend to develop towards the front edge of the landslide in the horizontal direction and extend towards the slip zone in the longitudinal direction. The soil behind the pile has completely undergone shear failure and entered the state of residual stress. The number of cracks on the surface of the front slope body of the pile increases further, resulting in tensile cracking and staggered deformation at the slope foot, forming transverse arc tensile cracks, and a small amount of soil sliding, and the deformation of the landslide front edge of the threaded pile side is more obvious (see Figure 8(c)).

When the thrust load reaches 4 MPa, the height of the soil uplift at the back edge of the landslide increases sharply, and the swelling crack behind the pile expands obviously. The number of cracks on the side of the sliding body behind the pile increases sharply, forming through cracks in the longitudinal connection, resulting in most of the soil being squeezed out by upward sliding with an angle of about 30° between the sliding direction and the horizontal direction. The transverse fractures continue to develop towards the front of the landslide and connect with the front tension fractures. The surface of the front slope of pile has formed a continuous pull-crack and staggered fall, and a large amount of soil has extruded and slipped, and the number and length of cracks on the pile top have increased obviously. At this point, the antislide pile completely failed and could not provide enough resistance, and the whole model of landslide-pipeline system reinforced by micropiles had formed an overall failure (see Figure 8(d)).

Based on the above macroscopic deformation signs of the landslide model, the landslide deformation process under the reinforcement of micropiles can be divided into four stages under the action of the thrust force of the landslide rear edge: tensile cracking deformation of the landslide rear edge → shear deformation caused by stress concentration of rock and soil in front of the pile → deformation failure of micropiles → tensile cracking failure of the landslide front. Compared with the engineering site before micropile reinforcement, it can be found that the deformation process of landslide under micropile reinforcement is similar to that in the engineering site, but the development of slip process is effectively inhibited. Although the landslide has a large deformation under the reinforcement of micropile, it does not slide down as a whole, which indicates that micropile has a good effect on the reinforcement of landslide.

4.1.2. Analysis of Horizontal Displacement of Pile Top

Micro-antislide pile is a concealed structure deeply embedded in rock and soil. As a macroindex to judge whether antislide pile fails, pile deformation, especially pile top displacement, can be used as an important condition to analyze the deformation and failure of antislide pile [29]. According to the phased characteristics of horizontal displacement curve of pile top under different loads in Figure 9, the deformation process of micro-antislide pile under landslide thrust can be divided into three stages: initial deformation stage, uniform deformation stage, and accelerated deformation stage. Under the small external force, the stress distribution of the rock mass at the back edge of the landslide changes, resulting in tensile cracking deformation. In the process of the soil stress which continues to transfer to the front edge of the landslide, the antislide pile plays an effective role in retaining and resisting most of the landslide thrust. At this stage, the antislide pile only has a small displacement deformation. Then, the antislide pile enters the uniform deformation stage, and the pile top displacement increases linearly with the increase of landslide thrust force. When the landslide thrust exceeds the ultimate resistance of the antislide pile, the deformation of the micropile enters the accelerated stage, during which the pile structure begins to bend and crack, resulting in the overall failure and loss of retaining capacity.

Comparing the variation of horizontal displacement of pile tops of different micropiles on both sides, it can be seen that the variation of pile top displacement of micropiles on both sides with increasing load shows a high degree of consistency, and both piles reach the ultimate bearing capacity under the external force of 3 MPa. Under the same landslide thrust, the horizontal displacement of the pile top on the screw pile side is significantly higher than that on the perforated pipe side. During the accelerated deformation phase, the horizontal displacement of the top of the screw pile tends to increase in a precipitous manner, while the horizontal displacement of the top of the perforated pipe increases at a slower rate than that of the screw pile. This indicates that even though the perforated pipe reaches the ultimate load carrying capacity under the action of landslide thrust, there is still a strong residual resistance to further deformation of the landslide.

4.2. Analysis of Stress and Deformation Characteristics of Pile
4.2.1. Distribution of Earth Pressure around Piles

Under the action of landslide thrust, the internal force variation and distribution characteristics of rock mass before and after antislide pile can reflect the action effect of antislide pile well. Therefore, in this test, earth pressure sensors were installed in front and back of two different micropile groups along the depth direction to monitor the stress changes of rock mass in front and back of the piles. Soil pressure distribution in front and back of pile under various working conditions is shown in Figure 10.

From Figure 10(a), it can be seen that the overall soil pressure on both sides of the perforated micropile is in the form of slide layer > slide zone > bedrock layer, which is basically consistent with the actual “pushed landslide” thrust load transfer law. With the increase of landslide thrust load, the soil pressure on the front and back sides of the pile also increased and showed a significant increase after working condition 5 (pressurized load 3 MPa), which was consistent with the change trend of pile top displacement mentioned above. At this time, the pile has reached the horizontal ultimate bearing capacity. By comparing the soil pressure values on the front and back sides of the pile, it can be seen that the soil pressure on the back side of the pile is significantly higher than that on the front side of the pile under various working conditions, indicating that the splined pipe pile plays an effective resistance role under the action of landslide thrust and offsets part of the soil stress. On the other hand, from the perspective of the change curve of earth pressure along the pile depth direction, the distribution law of earth pressure on both sides presents height consistency, which is approximately “S” curve along the pile depth distribution diagram, but the peak earth pressure depth on both sides appears in different positions. The peak earth pressure at the pile depth of 0.4 m is 46.09 MPa at the back side directly affected by landslide thrust. Due to the interaction of residual thrust and resistance of sliding body in front of pile, the peak earth pressure at 0.6 m pile depth is 32.57 MPa.

It can be seen in Figure 10(b) that the distribution law of earth pressure on both front and back sides of the screw micropile is similar to that of the perforated pipe micropile. The thrust load of landslide is still the fundamental reason for the change of soil pressure on both sides of pile, especially the increasing law of soil pressure behind pile with thrust load is obvious. Under the working condition 7 (thrust load 4 MPa), the earth pressure in front of the threaded micropile has a great mutation near the position in the pile. At this time, the middle part of the threaded micropile has been damaged and cannot bear the landslide thrust, resulting in a large increase in the stress of the sliding body in front of the pile. According to the research results of existing scholars, the earth pressure above the sliding surface is caused by the transfer of thrust load, and the earth pressure below the sliding surface is mainly caused by the deformation of micropile group [30]. It can be inferred that the thrust load of the threaded pile is the largest at the upper part of the pile above the sliding surface, and the deformation is the largest at the bottom of the pile below the sliding surface.

4.2.2. Analysis of Pile Bending Moment

The single pile in the first row, the single pile in the second row, and the single pile in the last row are selected from the perforated pipe micropile group and the screw micropile group, respectively, as the research object. According to the strain data collected on both sides of the pile, bending moment values of each section of the pile can be calculated by the following formula [31]: where is the bending moment of the test section (kN·m), is the elastic modulus of pile material (MPa), is the moment of inertia of the test section (m4), is the pile strain behind the micropile, is the pile strain in front of the micropile, and is the pile diameter (m). Figures 11 and 12 show the distribution of bending moment at measuring points of each section of the two piles under graded loading. The positive bending moment in the figure represents the tension on the rear side of the pile.

It can be seen from the figure that with the increase of external force of landslide, the bending moment value of splined pipe pile shows a hierarchical growth. The bending moment of the pile body of the first row presents an inverted “M” distribution along the depth direction, the bending moment of the pile top and bottom is small, and the bending moment of the pile body is positive only near the sliding surface, while the maximum negative bending moment of the pile body is -71.9 kN·m at the position about 1/4 of the pile length. The distribution form of bending moment of the second row pile is quite different from that of the front pile. In particular, the stress characteristics of the pile in the antisliding section change, and the deformation of the back side of the pile changes from compression to tension. The maximum positive bending moment value is 56.5 kN·m. From the distribution form of bending moment of pile body, the stress deformation of the final row pile is consistent with that of the first row pile, but the serious deformation of the final row pile is mainly concentrated in the antisliding section and the vicinity of the sliding surface, and the deformation increases sharply at the condition 7 (when the load is 4 MPa), and the maximum negative bending moment value reaches -106.1 kN·m, while the deformation of the anchoring section is small. Based on the bending moment distribution of the single pile at different locations in the group pile and the above-mentioned soil pressure distribution characteristics of the pile perimeter, it can be inferred that the perforated micropile is firstly deformed and damaged in the antislip section of the pile under the landslide thrust, and the most severely damaged section is located near 1/4 of the pile length, while the anchorage section is less deformed and effectively plays the role of anchorage. On the other hand, according to the comparison of bending moment of single pile at different positions, its failure degree is as follows: the final row pile > the first pile > the middle pile. Therefore, attention should be paid to the design of perforated pipe micropiles. Micropiles are slender structures, and the single piles are arranged and combined into a whole antisliding structure through pile cap. Due to the limitation of pile cap and anchoring section on displacement, the antisliding section is prone to bending deformation under the action of landslide thrust. Under the action of landslide thrust, the landslide body behind the pile produces plastic deformation and basically loses its sliding resistance [32]. The landslide thrust is transmitted to the front pile first, and the sliding section of the first row of piles plays a resistance role to resist part of the landslide thrust. When the landslide thrust exceeds the ultimate bearing capacity of the pile, the middle of the antislip section of the pile starts to bend and deform, while the internal force of the pile gradually shifts from the antislip section to the anchorage section. With the increase of the landslide thrust, the middle of the anchorage section also starts to show large deformation, which eventually leads to the failure of the overall pile bearing capacity due to the excessive deformation of the middle of the antislip section.

It can be seen from Figure 12 that, under the action of landslide thrust, the bending moment distribution forms of the first row, the middle row, and the last row of the threaded micropile group are highly similar, indicating that the deformation of each single pile in the threaded micropile group is synchronous. The bending moment distribution of each single pile presents an “S” shape along the depth direction. The positive and negative bending moment positions of the pile are roughly rotationally symmetric near the simulated sliding surface. Most sections above the sliding surface are negative bending moment, while the positive bending moment is below the sliding surface. The largest deformation degree of the first row pile is concentrated in the upper part of the pile and near the sliding surface. The maximum negative bending moment is 72.3 kN·m at the pile depth of 0.4 and 65.8 kN·m at the pile depth of 0.9 m (near the simulated sliding surface). The maximum deformation of the middle pile is at the lower part of the antisliding section, and the maximum bending moment is 84.6 kN·m. The deformation degree of the final row pile is the largest in the upper part of the anchoring section, and the maximum bending moment is 80.5 kN·m. Because the sliding section of the front pile first bears the landslide thrust transmitted by the back edge of the landslide, the bending moment of the sliding section of the middle pile and the back pile decreases obviously, but the deformation of the upper part of the anchoring section increases obviously, indicating that the anchoring section of the threaded pile is also a weak point of sliding resistance in practical application. In addition, under the same external force of landslide, the overall deformation degree of the screw micropile is greater than that of the perforated pipe micropile, and its sliding resistance is relatively weak. Therefore, attention should be paid to the use of the screw micropile in landslide antisliding design.

4.3. Analysis of Stress and Deformation Characteristics of Pipeline
4.3.1. Stress Distribution of Slip Body on Mountain Side and River Side of Pipeline

In order to study the deformation characteristics of natural gas pipeline under the action of micropiles, the soil stress changes on both sides of the pipeline are monitored by embedding soil pressure boxes. Figure 13 shows the distribution of peak earth pressure at each measuring point on the mountain side and river side of the pipeline. The TG1/TG1 measuring point of section I is located at the support side of perforated pipe micropile, the TG3/TG3 measuring point of section III is located at the support side of threaded pile, and the TG2/TG2 measuring point of section II is located at the pipeline interface.

As shown in Figure 13, the earth pressure at each measuring point on the mountain side and river side of the pipeline increases as the load increases. The distribution law of soil pressure on the side of the pipeline is as follows: section III > section II > section I, and under working condition 6 (loading 3.5 MPa), the soil pressure on the side of the pipeline mountain increases by leaps, and the peak earth pressure at TG3 measuring point reaches 22.99 kPa, which is nearly two times higher than that on the supporting side of the perforated pipe micropile on the same side. At this time, the pile has reached the ultimate bearing capacity, but there is a large difference in the earth pressure around section I and section III. It is further verified that the above pipe pile can still play a certain resistance after reaching the ultimate bearing capacity. Due to the bearing capacity of the pipeline itself and the resistance of the river side soil, the distribution of soil pressure on the river side of the pipeline changes, which is numerically expressed as section II > section III > section I; TG2 point near the pipeline maintains a high earth pressure, indicating that the pipeline has lost its own resistance at the interface, resulting in an increase of earth pressure transmitted to the river side of the pipeline. The TG3 measurement point located at section III has a sharp increase in earth pressure under the loading pressure of 4 MPa. At this time, the pipeline buried in front of the microthreaded pile has been completely destroyed, and the landslide begins to slip and deform along the sliding surface as a whole. Although the earth pressure at TG1 measurement point at section I has an upward trend, it remains in a stable section as a whole, and the perforated pipe micropile still has a certain protective effect on the pipeline.

4.3.2. Dynamic Evolution Process of Pipeline Deformation

In order to monitor the deformation characteristics of the pipeline under the reinforcement of micropiles caused by landslides, the dynamic deformation process of the pipeline was deduced by picking up the real-time strain signals of different sections of the pipeline with loading changes by strain sensors. Figure 14 shows the strain time history curves of measuring points in each section of the pipeline.

It can be seen from Figure 14 that the strain of the measured point on the mountain side of the pipeline is negative, while that on the river side is positive. The bending deformation of the open surface of the convex landslide front edge of the pipeline occurs, and finally, the tensile fracture failure occurs on the river side of the pipeline. The deformation process of pipeline can be divided into four periods (stable period, initial deformation period, constant velocity deformation period, and severe deformation period) according to the stage characteristics of the strain time history curve of the measuring point. When the thrust load is 1-2 MPa, the strain value of the pipeline is near zero. At this time, the pipeline is in a stable phase, and there is almost no deformation or small deformation in some parts. When the loading increased to 2-2.5 MPa, the strain value of the pipeline increased slightly, and the pipeline was in the initial deformation stage. When the loading reaches 2.5-3.5 MPa, the pipeline strain increases linearly and the curve slope is small. At this time, the pipeline reaches its ultimate bearing capacity and enters the constant deformation stage. When the thrust load increases to 4 MPa, the strain value at the position of the pipeline interface rises linearly, and the pipeline deformation enters the stage of intense development. At this time, the pipeline interface is first cracked and destroyed and is torn at both ends along the position of the interface [33]. After that, the pipeline is basically completely damaged and cannot be operated normally. Therefore, in the design and construction of the emergency treatment engineering of pipe-landslide system, it is necessary to strengthen the protection of the position of the pipeline interface, strengthen the bearing capacity of the pipeline interface section, and ensure the safety of the pipeline.

5. Discussion of the Results

5.1. Comprehensive Analysis of Multisource Data

In order to study the synergistic mechanism of landslide, micropile structure, and pipeline structure under the action of external forces, multivariate data of typical measuring points were selected for fusion analysis on the side of perforated pipe micropile and the side of screw pile, and the time history curve of data of each measuring point was drawn as shown in Figure 15.

It can be seen from Figure 15 that, for the two different types of root pile reinforcement pipeline-landslide system, the external force of soil landslide under the influence of the internal stress, the moved pile, pile strain, and pipe strain changes of the elements such as besides numerical size differences, on the changing trend is highly synchronized and coordinated. Therefore, the landslide rock mass and internal structure can be viewed as a whole when judging the instability mechanism of the whole system. According to the variation trend of each data factor at different time nodes and the above-mentioned deformation characteristics, the system deformation can be divided into four stages: initial deformation stage I, uniform deformation stage II, accelerated deformation stage III (divided into the first accelerated deformation stage (1) and the second accelerated deformation stage (2)), and residual deformation stage IV [34]. In stage I, under the influence of external force, the soil stress behind the pile begins to rise and is transferred to the micropile. After the micropile produces a small strain, its resistance rapidly develops. After that, the deformation of the pile body enters the elastic stage, at which the whole system remains stable. With the increment of external load, under the effect of root pile reinforcement, the deformation of the pipeline lag began to form the microdeformation in II phase, and Figures 15(a) and 15(b) show that under the effect of perforated pipe miniature pile supporting and relative delay of pipeline deformation, it can give the actual landslide deformation of pipeline accidents and increase the rescue time. When the loading lasted for 130 minutes, the external load reached 3.5 MPa, the pile reached the yield strength, and the system entered the first stage of accelerated deformation. The stress and structural deformation of soil inside the landslide began to jump but remained stable in a short time, and the pile was still exerting resistance. After a period of stability, the system enters the second accelerated deformation stage, during which the deformation of the system increases sharply, and the increase is more than twice that of the first accelerated deformation stage. The pile reaches the ultimate strength, resulting in serious bending deformation. After the failure of pile, the landslide thrust force rapidly compresses the pipeline. Due to the low strength of the pipeline, tensile fracture failure occurs in a short time, and then, the whole system is destroyed and enters the residual deformation stage. In addition, according to the comparison between Figures 15(a) and 15(b), it can be seen that the yield strength and ultimate strength of the perforated pipe micropile are both greater than those of the threaded pile, and it can still exert a certain residual resistance to keep the pile pressure stable after reaching the ultimate bearing capacity. The pressure grouting of the microtubular pile makes the pile body and the surrounding soil form a composite solid, which greatly enhances the sliding resistance of the single pile and effectively plays a synergistic role under the thrust of the landslide. On the one hand, it strengthens the landslide, and on the other hand, it plays a key role in the protection of the pipeline in front of the pile.

5.2. Analysis of Micropile-Pipeline Interaction

Under the action of landslide thrust, the loading section of the micropile mainly bears the landslide thrust behind the pile and the resistance of sliding body in front of the pile. After the resistance of the micropile, the pipeline mainly bears the residual landslide thrust in front of the pipe and the resistance of sliding body in front of the pipe, as shown in Figure 16. In order to compare the effect of perforated pipe micropile and screw micropile on pipeline protection, the sliding resistance coefficient of micropile and pipeline is defined as where is the average earth pressure of each measuring point in the load section of the micropile and is the earth pressure measuring point value behind the pipe. In other words, when the value is between 0 and 1, the larger the value is, the stronger the antisliding effect of the micropile is, the smaller the residual landslide thrust acting on the pipeline is, and the more prominent the protection effect of the micropile on the pipeline is.

Figure 17(b) shows the values of perforated pipe micropile and screw micropile under different thrust forces. It can be seen that the value of perforated pipe micropile is greater than 0.7 under different loads, while the value of screw micropile decreases with the increase of external thrust, even to below 0.1 under the action of 4 MPa thrust. It shows that perforated pipe micropile can effectively protect pipeline safety under different grades of landslide thrust, while screw micropile cannot effectively resist landslide thrust under large load level, resulting in greater risk of pipeline failure. By comparing the bending moment values of micropile and pipeline in Figure 17(a), it can be seen that the bending moment value of screw micropile is larger than that of perforated pipe micropile on the whole, and the bending moment value of pipe on the side of screw micropile is larger than that of perforated pipe micropile on the whole, which further indicates that the risk of pipeline failure under the support of screw micropile is greater, and perforated pipe micropile has a more prominent protection effect on pipeline.

6. Summary and Conclusions

The load-bearing mechanism of the two kinds of micropiles and the deformation characteristics of pipeline in front of the pile under the action of landslide were studied by large-scale laboratory model test, and the interaction and cooperative deformation mechanism of landslide, micropiles, and pipeline under the action of external force were analyzed. The preliminary analysis shows that the deformation process of landslide, micropile, and pipeline under external force is highly synchronized and coordinated and can be divided into four deformation stages: initial deformation stage, uniform deformation stage, accelerated deformation stage, and residual deformation stage. The main conclusions are as follows: (1)The peak stress of the soil behind the pile is concentrated in the middle of the antisliding section, and the peak stress of the soil behind the pile moves down to the vicinity of the sliding surface. The micropile is subjected to the strong shear force of the soil on both sides near the sliding surface(2)The deformation of the microtubular pile is serious in the middle of the antisliding section and near the sliding surface, and the bending moment reaches the peak from the top to the bottom quarter of the pile, and the bending failure occurs first. The deformation of the first row of piles and the last row of piles is synchronized, while the bending direction of the antisliding section of the middle pile is opposite. The deformation of every single pile in the screw microgroup is synchronous. The bending moment distribution of the pile body is “S” shape along the depth direction, and the positive and negative bending moments of the pile body are rotationally symmetrical on both sides of the simulated sliding surface. The bending failure of the first row of pile first occurs at 1/5 of the pile length, while the reverse bending deformation of the last row pile first occurs near the sliding surface. Attention should be paid to the treatment of weak areas of different piles in the design and construction of micropiles(3)The pipe in front of the pile only produces small deformation during the initial deformation stage and uniform deformation stage, and it can still operate normally during this period. Moreover, the support of the perforated pipe micropile delays the initial deformation stage of the pipe, which can increase the rescue time for the pipe deformation accident caused by the actual landslide. On the other hand, the pipe welding joint is the weak position of the antideformation of the pipe. After the failure of the pile, the pipe joint is first broken and torn along both ends and finally completely broken(4)Due to the pressure injection of the perforated micropile, the pile and the surrounding soil form a composite solid, which greatly enhances the slip resistance of the single pile and the effective synergy of each single pile under the landslide thrust. As a result, the yield strength and ultimate strength of the perforated micropiles are greater than those of the screw micropiles under the same working conditions, and they can still play a certain residual resistance role after reaching the ultimate bearing capacity. It has a beneficial effect on the reinforcement of pipeline crossing landslide system.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest

We declare that we do not have any commercial or associative interest that represents a conflict of interest in connection with the work submitted.

Authors’ Contributions

Wei Guan was responsible for the conceptualization, methodology, data curation, and writing of the original draft. Honggang Wu was responsible for the visualization, investigation, resources, project administration, supervision, and funding acquisition. Daoyong Wu was responsible for the conceptualization, methodology, validation, and formal analysis. Lin Tang was responsible for the formal analysis, validation, and conceptualization. Hong Wei reviewed and edited the manuscript. Honggang Wu contributed equally to this work and should be considered as co-first author.

Acknowledgments

This research was supported by the National Key R&D Program of China (No. 2018YFC1504904), the Natural Science Foundation of Gansu Province (No. 21JR7RA738), and the Science and Technology R&D Project of Southwest Pipeline Co., Ltd., State Pipeline Group (2020B-3106-0501).