Numerical Investigation and Prediction of Side-By-Side Tunneling Effects on Buried Pipelines

Round 1

Reviewer 1 Report

Comments for author File: Comments.pdf

Author Response

• C: In figure 2 and other figure, the capital letter X in the abscissa axis (Distance from tunnel centerline, X (mm) should be minuscules x.

R: Thanks for the comment. The capital letter X in the abscissa axis (Distance from tunnel centerline, X (mm) have been converted to minuscules x in the revised paper. To address this comment, the following revision is made: Fig 2.

Figure 2. Estimation of ground settlement of side-by-side tunnels.

 

• C: FEA: Mesh convergences should be involved, how many elements and nodes were used?

R: Thanks for the comment. PSI element is a particular pipe-soil interaction element in soft ware ABAQUS, which is often used in INP file. The mesh picture of pipe and soil is not often used. Hence, we added the the information about elements and nodes numbers in Fig. 3. To address this comment, the following revision is made:

 

Figure 3. Illustration of beam-on-spring FE model for tunnelling effect on pipeline: (a) idealized representation of soil with discrete springs, (b) transverse vertical spring, (c) axial spring.

Lines 154-157: “The pipeline is modeled using 1200 numbers of beam (PIPE31). The pipe-soil interac-tion is reflected by the compression and extension of soil springs along the pipeline, which is characterized by 1200 numbers of pipe-soil interaction (PSI) elements.”

 

• C: Figure 7-8,10-11, why the Longitudinal pipe stain is considered? while the upward, downward are not considered? The strains of these two directions are more important to the bending of pipelines.
  
  

R: Thanks for the comment. The longitudinal pipe stain is the measured strain in the existing pipeline along its longitudinal direction, με, which is common used to characterize pipeline response to bending. The longitudinal pipe stain is also used in Wang et al. (2011) and Ni and Mangalathu (2018) to express the characterize pipeline response subjected to tunneling induced ground settlement.

 

• C: The position of tunnels T1 and T2 in all figures should be unified.

R: Thanks for the comment. The position of tunnels T1 and T2 in all figures have been unified. To address this comment, the following revision is made:

Figure 1. Schematic view of side-by-side tunnelling effect on buried pipeline.

Figure 2. Estimation of ground settlement of side-by-side tunnels.

 

• C: The material of pipeline should be defined.

R: Thanks for the comment. The material of pipeline has been defined. To address this comment, the following revision is made:

 

Lines 176-180: “In the test, an aluminum alloy pipeline with an outer diameter of 0.635 m (in prototype scale), length of 36.8 m, burial depth of 1.2 m, wall thickness of 0.066 m, elastic modulus of 70 GPa was considered. The tunnel had an outer diameter of 6.08 m and burial depth of 12.04 m. The ground volume loss was about 2%. The soil friction angle is 31°, and unit weight is 15.19 kN/m³.”

 

REFERENCE

Ni, P., and Mangalathu, S. 2018. Fragility analysis of gray iron pipelines subjected to tunneling induced ground settlement. Tunnelling and Underground Space Technology, 76: 133-144. doi:10.1016/j.tust.2018.03.014.

Wang, Y., Shi, J., and Ng, C.W. 2011. Numerical modeling of tunneling effect on buried pipelines. Canadian Geotechnical Journal, 48(7): 1125-1137.

 

Reviewer 2 Report

This manuscript presents a numerical investigation and prediction of side-by-side tunnelling effects on buried pipelines. The authors propose a semi-empirical function to predict the settlement of side-by-side tunneling system. By going through the whole manuscript, the accuracy of the numerical model is doubted. The authors are recommended to consider the following the comments.

 

1.     Please check the manuscript thoroughly to correct the grammatical errors (e.g., the results of which is plotted in Fig. 9)

2.     What is the evidence to support K=0.3 and Dr=0.7 in the numerical analysis?

3.     Where are the input parameters in Table 1 from?

4.     The authors should re-write the numerical validation to highlight the logic to validate the accuracy of the numerical model.

Author Response

• C: Please check the manuscript thoroughly to correct the grammatical errors (e.g., the results of which is plotted in Fig. 9)

R: Thanks for the comment. We have revised the manuscript for the grammatical errors, and the revised contents have been marked yellow. To address this comment, the following revision is made:

Figure 9. Comparison of numerical results against dimensionless plot.

 

• C: What is the evidence to support K=0.3 and Dr=0.7 in the numerical analysis?

R: Thanks for the comment. D_r is the relative density of soil, K is the dimensionless trough width parameter. According to the centrifuge test (Shi et al. 2016), the relative density of soil, D_r, is 0.7; K can be taken as 0.3 for the cohesionless soil, such as sand.

 

• C: Where are the input parameters in Table 1 from?

R: Thanks for the comment. The input parameters are calculated by Eq. 7, which is obtained from ALA standard (ALA 2005). To address this comment, the following revision is made:

Lines 181-183: “According to ALA standard (ALA 2005) and the calculation parameters mentioned above, the input parameters for soil springs can be obtained Using Eq. 7, which are listed in Table 1.”

 

• C: The authors should re-write the numerical validation to highlight the logic to validate the accuracy of the numerical model.

R: Thanks for the comment. We have revised the manuscript. To address this comment, the following revision is made:

Lines 187-201: “The variations of normalized ground surface settlement, S/D, and pipe bending strain, με, calculated using finite element model are shown in Fig. 4, accompanied by the results measured from centrifuge tests. As can be seen, both the maximum values of ground surface settlement and pipe bending strain appear in the centerline, and the values tends to zero when |x| increases further. The predicted normalized ground surface settlement is S/D=0.44%, which is quite close to the measured result of 0.43%. Besides, the ground surface settlement curves also match relatively well with each other. The numerical result of maximum longitudinal pipe strain occurs at the maxi-mum settlement position, with a value of 207 με, and the corresponding measurement is about 200 με. The numerical results of maximum negative strain are smaller than that of the measurements, but the general shapes of strain curves are relatively close to each other. In general, the finite element method provides a good approximation to the measured values for the pipe-tunnel interaction centrifuge test. Note that only a single tunnel is considered here.  Therefore, the numerical model is extended to reflect the twin tunnelling effects on buried pipelines in the following sections.”

ALA. 2005. Guidelines for the design of buried steel pipe. American Lifelines Alliance, New York, NY, USA.

Shi, J., Wang, Y., and Ng, C.W. 2016. Three-dimensional centrifuge modeling of ground and pipeline response to tunnel excavation. Journal of Geotechnical and Geoenvironmental Engineering, 142(11): 04016054.

Reviewer 3 Report

1. Original submission

Recommendation: accept after minor revision

 

2. Comments to editors

Manuscript Number: sustainability-2063611

Title: Numerical investigation and prediction of side-by-side tunnelling effects on buried pipelines

 

Overview and general recommendation:

This article presents an effectiveness prediction of the ground settlement under twin tunneling condition. The pipe response under twin tunneling effect is then investigated by a series of beam-on-spring finite element analysis. The ground deformation and the corresponding pipe response under twin tunneling effects are more significant than those under single tunnel conditions. Subsequently, the effects of burial depth and twin-tunnel space are discussed. A semi-empirical prediction method for the longitudinal pipe bending strain was proposed. The paper is in good written, and the numerical investigation is quite helpful for tunnel and pipe engineers. I would like to recommend this paper to be accepted after minor revisions.

 

Specific comments:

1 Page 8: An error has occurred at the subtitle in Figure 3, where (d) and (c) are not correctly expressed. Page 9: An error has occurred at the subtitle in Figure 4, where (a) and (b) are not expressed. Therefore, the expression in Figs. 3 and 4 should be revised.

2 Page 5: Background about ground settlement is limited and needs to expand covering past research on ground settlement.

3 Page 5: There must be some assumptions about the theory of ground settlement, which should be expressed at first to understand the major limitations. Therefore, related assumptions should be added in revised manuscript.

4 Page 10: Table for Parametric analyses is needed in “4 Parametric analyses of twin-tunnelling effects on buried pipelines”, and the analysed parameters should be expressed separately.

Author Response

• C: Page 8: An error has occurred at the subtitle in Figure 3, where (d) and (c) are not correctly expressed. Page 9: An error has occurred at the subtitle in Figure 4, where (a) and (b) are not expressed. Therefore, the expression in Figs. 3 and 4 should be revised.

R: Thanks for the comment. We have revised the manuscript. To address this comment, the following revision is made:

Figure 3. Illustration of beam-on-spring FE model for tunnelling effect on pipeline: (a) idealized representation of soil with discrete springs, (b) transverse vertical spring, (c) axial spring.

 

Figure 4. Comparison with test measurements: (a) ground settlement, (b) longitudinal pipe strain.

 

• C: Page 5: Background about ground settlement is very much limited and needs to expand covering past research on ground settlement.

R: Thanks for the comment. We have revised the manuscript. To address this comment, the following revision is made:

Lines 99-101: “Tunneling induced ground volume loss can lead to the differential settlement of buried pipelines. The above analytical solutions rely predominantly on the use of greenfield settlements as an input in order to determine the pipeline’s generated re-sponse.”

 

• C: Page 5: There must be some assumptions about the theory of ground settlement, which should be expressed at first to understand the major limitations. Therefore, related assumptions should be added in revised manuscript.

 

R: Thanks for the comment. The assumptions about the theory of ground settlement have been added in revised the manuscript. To address this comment, the following revision is made:

Lines 101-104: “It is suggested here that a Gaussian curve (Peck 1969) can be used to fit not only the greenfield input but also the deformed pipeline. With regard to the greenfield settlement at the ground surface, a variety of predictive methods have been established”

 

• C: Page 10: Table for Parametric analyses is needed in “4 Parametric analyses of twin-tunnelling effects on buried pipelines”, and the analysed parameters should be expressed separately.

R: Thanks for the comment. We have added the table for parametric analyses, in which the analysed parameters expressed separately. To address this comment, the following revision is made:

Table 2. Summary of parameters

Item	Parameter	Value
Tunnel	Burial depth, Z0	2.0D, 4.0D
	Space, d’	1.0D, 1.5D, 2.0D
	Diameter, D	6.08 m
Pipe	Diameter, Dp	0.635 m
	Wall thickness, tp	0.066 m
	Burial depth, H	1.2 m
	Young’s modulus, Ep	70 Gpa
Soil	Friction angle, φ	31°
	Unit weight, γ	15.19 kN/m3

 

Round 2

Reviewer 2 Report

The manuscript can be accepted in current format.