Abstract
With the rapid increasing mileage of China’s long-distance oil and gas pipelines, the intersection construction with other civil engineering, encroachment by invading the pipeline route, and other third-part problems occurred occasionally. Heavy vehicles rolling the serving pipeline have become one of the inevitable safety risks. The field test method of three in-service long oil and gas pipelines under heavy vehicle rolling is elaborated. The pipeline stress and soil pressure during tests are monitored and analyzed. The parametric model of pipe mechanical response under heavy vehicle rolling and stacking load is established, and the quantitative feedback of each parameter on pipeline stress safety response under heavy vehicle rolling is formed. The results show that the influence factors such as vehicle weight, vehicle speed, buried depth of pipeline, horizontal distance and cover plate on the stress state, vertical displacement state, and contact stress of pipe soil interface are highly consistent with the finite element method and test method, which could be a guideline for the safety assessment at rolling site.
1. Introduction
By the end of 2020, the total mileage of national oil and gas pipeline will exceed 150 thousand kilometers. With the completion and commissioning of the fourth Shaanxi-Beijing gas pipeline, China’s pipeline network pattern of “west to east gas pipeline, north to South gas pipeline, Sea gas landing and nearby supply” has been basically taken shape [1]. The long-distance oil and gas pipelines crossing East to West and North to South and connecting overseas are inevitable to intersect with roads. The loads borne by the intersecting pipelines mainly include static loads, such as pipeline structure weight and soil pressure, and dynamic loads, such as vehicle rolling loads, ground stacking loads, temperature loads, and pipeline internal pressure, among which the most important one is vehicle loads. According to the investigation of relevant departments, overload vehicles have become the main cause of damage to civil structures in some areas. Vehicles do not directly roll on the pipeline, but the transmission of vehicle rolling load through the road and soil will eventually affect the safe operation of the pipeline [2–4].
At present, casing or cover plate (culvert) is adopted to protect the pipeline passing through the road in our country, but the direct laying method is still adopted in farmland and wasteland without additional protection. With the improvement of farmland and the change of land use, great changes have taken place in the environment around the pipeline. The initial road surface has been formed above the natural gas pipeline originally laid according to the requirements of farmland burial depth, and even heavy or overloaded vehicles will pass by, which is bound to bring serious challenges to the safe operation of the pipeline. Moreover, natural gas is flammable, explosive, and easy to spread. Once an accident occurs in a gas pipeline, it will not only cause direct losses to the pipeline itself and reduce the economic benefits of the pipeline companies but also cause a huge security threat to the residents and factories around the pipeline and bring serious property losses to the society. Based on this, the test on the influence law of heavy vehicle rolling on the actual pipeline safety in service is carried out, and the monitoring data are analyzed, forming the influence law of heavy vehicle rolling on the pipeline [5–8].
Based on the experimental study on the response and mechanical characteristics of buried pipeline under vehicle rolling load, Meng et al. and Alzabeebee et al. established an experimental model of shallow buried steel pipeline and launch a comparative study of the response. The experiment shows that the bending strain of pipeline decreases with the increase of axle load and vehicle speed, and the wheelbase has little effect on the maximum strain of pipeline [9, 10]. Cao et al. and Shin and Lee measured the vertical earth pressure in different depth of foundation soil caused by vehicle rolling load and obtained the relationship between the duration of earth pressure and vehicle speed. Through the soil pressure test of heavy vehicle rolling [11, 12], Feng et al. summarized the distribution law of soil stress caused by vehicle load and established the finite element model of heavy vehicle rolling. Through the analysis of the result data, they obtained the minimum buried depth to ensure the safety of pipeline operation [13]. These experimental studies are not carried out for the actual pipeline with pressure, and there is a large deviation from the real working conditions.
In the finite element analysis method, Li and Zhang et al. simplified the vehicle load model into three forms: long-term uniform load, moving dead load, and steady-state harmonic load and used the ABAQUS software to analyze the mechanical characteristics of urban water supply pipeline under various simplified models. The stiffness effect of internal pressure and the support of soil at the sides of the pipe are also studied considering both hoop stress and longitudinal stress resulting from surface loading [14, 15]. Lan et al. and Li and Chen analyzed the soil propagation and movement law of wave caused by vehicle load. In this paper, the coupling problem of buried pipeline and soil under vehicle load is studied, and the finite element model is established to analyze the factors that may cause pipeline failure [16, 17]. Guo et al. and Khademi-Zahedi used the ANSYS software to establish the finite element model of buried PE pipeline, calculated the Mises stress and diameter deformation of pipeline under different vehicle loads, and studied the influence of vehicle loads on PE pipeline [18, 19]. The result analysis of these similar projects has no applicability and operability and can not be directly applied to the pipeline operation site for on-site safety assessment.
There is no unified method for the study of dynamic response of buried pipeline under vehicle rolling load at home and abroad. Scholars generally focus on the stress of buried pipeline under vehicle rolling load. The research methods are mainly divided into formula calculation method, experimental method, and numerical simulation method. Vilkys et al., Rusin et al., McDonough, and Ksenofontov divided the load of buried pipeline into fill load and surface traffic live load and calculated the vertical pressure of buried pipeline under the action of surface traffic live load by using Boussinesq point load method, Spangler method, distribution angle method, and Newmark integral superposition method based on Boussinesq equation [20–23]. However, several methods are not combined with the real feedback law of in-service pipeline. There are few previous studies on stress prediction method of buried pipeline under heavy vehicle load. To fill this gap, experimental and numerical investigations on stress response prediction for buried pipeline are proposed.
2. Experimental Investigation
In light of the theoretical feedback of heavy vehicle rolling on operating pipeline, we had carried out heavy vehicle rolling field experiments for three times on the operating pipelines of North PipeChina Company: (1) a diameter of 660 mm product oil pipeline, (2) a diameter of 1016 mm natural gas pipeline, and (3) a diameter of 711 mm natural gas pipeline in which 2 MPa inner pressure was operating.
The first two field tests placed two monitoring sections, sections A and B, and the third field test only set one detection section, where section A represents the strain monitoring section just under the place of heavy vehicle rolling and section B represents the strain monitoring section about 5 ~ 6 m away from section A.
Each monitoring section has three monitoring points: 0°, 90°, and 270° sensors are installed, and each point is equipped with two sensors. In the diagram, stands for 0° position, stands for 90° position, and stands for 270° position; 1 represents that the sensor is installed along the axial direction of the pipeline, and 2 represents that the sensor is installed along the circumferential direction of the pipeline.
The first heavy vehicle rolling test was carried out in Mengjin, Henan Province. The field test used a diameter of 660 mm product oil pipeline which is abandoned after the transport line changed. The design pressure was 8-14 MPa, the actual operating pressure was 2-6 MPa, the outer diameter of the pipe was 660 mm, the material was X65, and the wall thickness was 12.7 mm. The buried depth of the pipe top is 2.6 m, and there is no internal pressure during the experiment. The rolling vehicle is DELONGXIN M3000, dump truck, and container size . The empty weight is 12.5 tons, the full load earthwork is 20 cubic meters, the earthwork weight is 36 tons, and the total weight is about 48.5 tons. The second and third tests are similar to the first test scheme, and the site is similar (Figure 1).
(a) Sensor installation site
(b) On-site rolling of buried pipeline with heavy vehicle and full load
The process of the experiment is as follows: first, the empty vehicle is rolled back and forth; then, the half loaded vehicle is rolled back and forth; then, the full load is rolled back and forth; finally, the cover plate is added, and then, the full load is rolled back and forth [24–32].
The time-strain data of four experimental conditions, i.e., empty car rolling, half car rolling, full load rolling, and full load rolling with cover plate rolling, are extracted. Along the axial direction of the pipeline, the 12 o’clock direction is the upper surface of the pipeline, and the 3 o’clock and 9 o’clock directions are the side of the pipeline. The actual data obtained are shown in Figure 2.
(a) Time strain curve of empty vehicle rolling back and forth
(b) Time strain curve of half vehicle rolling back and forth
(c) Full load back and forth rolling time strain curve
According to the experimental data, the following conclusions can be drawn: (1)The circumferential strain at 90° is the largest, followed by that at 270°(2)The axial strain of the pipeline is basically 0, and the variation range is very small(3)The strain of the strain monitoring section at the place of heavy vehicle rolling is greater than the strain monitoring section about 5-6 m away from the section(4)The greater the vehicle weight is, the greater the strain is. The cover plate can effectively reduce the strain of the pipeline(5)When the loaded truck passes the vertical pipeline at a certain speed, the pipeline is subjected to a considerable positive pressure at 0°, resulting in elliptical deformation. Therefore, the large strain of the pipeline is mainly concentrated in the ring direction of 90°and 270°. Because of the restriction of soil, the axial direction of pipeline will not deform
3. Rolling Law of Heavy Vehicle Based on Finite Element Method
Due to the effective data obtained on site and the limited working conditions of field test, in order to study the influence of internal pressure, pipe diameter, wall thickness, body weight, site soil characteristics, vehicle speed, pipeline buried depth, horizontal distance, and other factors on the pipeline, the finite element method can be used to realize data expansion.
3.1. Finite Element Modeling
Based on the feedback law of heavy rolling pipeline, the field test is studied. The numerical simulation model of the mechanical response of the pipeline under rolling load is established by using ABAQUS. The inversion research is carried out for the experimental conditions. The geometry model is shown in the following figures. The whole finite element model includes subgrade, site soil, buried pipeline, and cover plate. The soil area of the site is 21 m long, 12 m wide, and 5 m high. The outer diameter of the pipe is 1016 mm, and the wall thickness is 21 mm. The buried depth of the pipe top is 1.6 m. The thickness of cover plate is 20 mm.
Because the wall thickness of the pipe is obviously smaller than the axial length of the pipe, the shell element simulation can be used, and the element type is set to four-node curved shell unit S4R. The element type can be used for modelling thin or thick shell structures. The reduced integral method, including hourglass mode control, allows the finite film strain to simulate the stress and strain response of the pipeline better. The soil, subgrade, and cover plate of the site are simulated by solid element, and the element type is 8-node hexahedron linear reduction integral unit C3D8R. The results of linear reduction integral element are more accurate. In order to accurately retrieve the feedback response of buried pipeline under rolling action of vehicles, local grid encryption is carried out for the pipeline and subgrade in the model. The whole model contains 325198 (Figure 3).
(a) Finite element modeling of heavy vehicle rolling
(b) Loading mode of vehicle load
3.2. Analysis of Pipe-Soil Interaction
The M-C (Mohr-Coulomb) model and D-P (Drucker-Prager) model can be used in the ABAQUS software to study the stress safety feedback response of pipeline under external load [16]. On the basis of the D-P model, the yield surface of M-C criterion on plane is changed to smooth surface, and the original material model is improved accordingly. Therefore, the D-P model is selected as the material model of soil in this project, and the expression of the D-P model is shown in
3.3. Numerical Inversion Investigation
In order to accurately invert the feedback response of buried pipeline under vehicle rolling load under test conditions, the gravity of pipeline and soil and the internal pressure of pipeline are considered in the numerical simulation model. According to different experimental conditions, the internal pressure of the pipeline is set as 2 MPa and 0 MPa, respectively.
Under the joint influence of vehicle vibration and road roughness, the vehicle rolling load will show the characteristics of random variation. In order to accurately invert the mechanical response of buried pipeline under rolling load, the rolling load of heavy vehicle with random variation characteristics is decomposed into two dimensions of time and space. In the spatial dimension, it is necessary to determine the loaded vehicle pressure per unit area and the spatial position change of the loaded vehicle load during vehicle driving according to the contact area between the wheel and the ground and the driving speed of the loaded vehicle. The imprint of automobile tire is almost oval, and the pressure distribution on the imprint is not completely uniform. In order to facilitate the research, the project assumes that the rolling load of vehicles is evenly distributed on the contact surface. The actual contact area between each tire and the ground is calculated as a rectangular area. The total contact area between the vehicle and the ground is calculated to be 0.84 m2.
In the time dimension, the contact pressure between the wheel and the ground changes randomly with time. Combined with the field measured data, the variation curve of vehicle with time is fitted.
ABAQUS provides a large number of user subroutines for situations where ABAQUS cannot solve directly. Through these subroutine interfaces, users can define boundary conditions, load conditions, contact conditions, and material properties as needed, which greatly expands the availability and flexibility of ABAQUS [3, 4, 33]. Based on the secondary development method of FORTRAN language and DLOAD user subroutine, this paper completes the application of model heavy vehicle rolling boundary conditions. Accurately consider the variation characteristics of heavy vehicle load with time and space.
The variation curve of the circumferential strain of the pipeline in the 90° direction at the monitoring surface of the heavy vehicle rolling back and forth with the driving time of the heavy vehicle is extracted (Figure 4).
(a) Arrangement position of strain gauge under rolling load of heavy vehicle
(b) Comparison curve of strain with time
4. Analysis on Influencing Factors of Pipeline Safety State under the Traffic Load
4.1. Effects of Vehicle Speed
The influence of vehicle speed on the safety state of pipeline under heavy vehicle rolling is analyzed below. Taking the benchmark working condition as the standard, considering the vehicle driving speed of 1 m/s, 2 m/s, and 3 m/s, parallel calculation is carried out.
The time corresponding to the maximum equivalent stress is obtained through the variation curve of the equivalent stress of the dangerous section of the pipeline with time during heavy vehicle rolling. Based on the numerical calculation results, the axial variation trend of axial stress in the pipeline under different vehicle speeds is extracted as follows. It can be seen that with the increase of vehicle speed, the axial stress at the top, bottom, and side of the pipeline increases (Figure 5).
(a) Pipe crown
(b) Pipe springline
The axial variation trend of circumferential stress under different vehicle speeds is extracted as follows. With the increase of vehicle speed, the circumferential stress at the top, bottom, and side of the pipe increases (Figure 6).
(a) Pipe crown
(b) Pipe springline
4.2. Effects of Vehicle Weight
The influence of body weight on the safety state of pipeline under heavy vehicle rolling is analyzed below. Taking the benchmark working condition as the standard and considering the body weight of 40 t, 50 t, 70 t, and 90 t, parallel calculation is carried out.
The time corresponding to the maximum equivalent stress is obtained through the variation curve of the equivalent stress of the dangerous section of the pipeline with time during heavy vehicle rolling. Based on the numerical calculation results, the variation trend of Mises stress at the pipe top under different body weight is extracted as follows. It can be seen that the change of body weight has little effect on Mises stress at the pipe top.
The axial variation trend of surface soil and pipeline displacement under different body weight is extracted as follows. It can be seen that the displacement of soil and pipeline increases significantly with the increase of body weight (Figure 7).
(a) Pipe crown
(b) Soil top
4.3. Effects of Pipe Diameter
The influence of pipe diameter on the safety state of pipeline under heavy vehicle rolling is analyzed below. Taking the benchmark working condition as the standard, considering the pipe diameters of 711 mm, 813 mm, 914 mm, and 1016 mm, parallel calculation is carried out.
The time corresponding to the maximum equivalent stress is obtained through the variation curve of the equivalent stress of the dangerous section of the pipeline with time during heavy vehicle rolling. Based on the numerical calculation results, the axial variation trend of axial stress under different pipe diameters is extracted as follows. It can be seen that with the increase of pipe diameter, the axial stress at the top, bottom, and side of the pipe increases significantly (Figure 8).
(a) Pipe crown
(b) Pipe springline
The axial variation trend of pipe soil interface pressure under different pipe diameters is extracted as follows. It can be seen that with the increase of pipe diameter, the pipe soil interface pressure everywhere in the pipeline increases, but the increase is not obvious (Figure 9).
5. Analysis of Safety State of Pipeline Rolled by Heavy Truck under the Influence of Factors
Based on the calculation results of the pipeline finite element model under heavy vehicle rolling obtained from the test and finite element method, the semiempirical prediction formula of Mises stress of the pipeline under heavy vehicle rolling is determined by using the nonlinear fitting method. The specific results are as follows:
where is the Mises stress, MPa; Speed is the vehicle speed, m/s; Thick is the pipe wall thickness, mm; Inpre is the internal pressure of the pipeline, MPa; and Vdis is the horizontal distance, m.
According to the results of pipeline Mises stress and soil displacement calculated by finite element method, the correlation between pipeline stress and soil displacement is studied, and the following results are obtained (Figure 10).
6. Conclusion
Based on the parameter analysis results of heavy vehicle rolling and surcharge, the sensitivity ranking of the indexes and influencing factors of pipeline mechanical response is carried out by the grey correlation analysis method. The results show the following.
The selected influencing factors with high sensitivity such as internal pressure, pipe diameter, and wall thickness are fitted by to obtain the fitting formula of pipeline mechanical response index, which can realize the comprehensive comparison of various influencing factors of pipeline. The error between the proposed fitting formula and the numerical simulation method is less than 10%, and it is also in good agreement with the actual monitoring value. Axial stress produced by heavy truck rolling load on X65 pipeline generally does not exceed 100 MPa. The established multifactor comprehensive pipeline safety classification evaluation method under the action of heavy vehicle rolling can realize multifactor sensitivity analysis and put forward corresponding treatment methods on the site according to different sensitivities. The ranking of sensitive factors can assist in site risk mitigation. After verification, the quantitative index and prediction method of the impact of heavy vehicle rolling on the pipeline proposed in this paper can greatly guide the safe operation of the pipeline on site and provide reference and help for the safe operation and daily management and maintenance of the pipeline.
Data Availability
All the data referred are shown in the figures and tables in the manuscript file.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
The research is supported by the General Institute of Science and Technology Company of PipeChina and the China University of Petroleum, Beijing.