See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/326618496

Response analysis of an aerial-crossing gas-transmission pipeline during

pigging operations

Article  in  International Journal of Pressure Vessels and Piping · July 2018

DOI: 10.1016/j.ijpvp.2018.07.011

CITATIONS READS

0 17

5 authors, including:

Lin Wang
Southwest Jiaotong University
8 PUBLICATIONS   10 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

MULTIPHSE FLOW View project

Stability of pipes conveying fluids View project

All content following this page was uploaded by Lin Wang on 03 August 2018.

The user has requested enhancement of the downloaded file.

 International Journal of Pressure Vessels and Piping 165 (2018) 286–294

Contents lists available at ScienceDirect

International Journal of Pressure Vessels and Piping

journal homepage: www.elsevier.com/locate/ijpvp

Response analysis of an aerial-crossing gas-transmission pipeline during T

pigging operations
Lin Wanga,∗, Haisheng Bib, Yiren Yangc, Yongxiang Zhangd, Yuxing Lie
a
School of Mechatronic Engineering, Southwest Petroleum University, Chengdu, 610500, China
b
College of Electromechanical Engineering, Qingdao University of Science & Technology, Qingdao, 266061, China
c
School of Mechanics and Engineering, Southwest Jiaotong University, Chengdu, 610031, China
d
China Petroleum Pipeline Engineering Corporation, Langfang, 065000, China
e
College of Pipeline and Civil Engineering, China University of Petroleum (East China), Qingdao, 266580, China

A R T I C LE I N FO A B S T R A C T

Keywords: Pigging operations of natural gas transmission pipelines for cleaning, swabbing and batching inhibition become
Natural gas pipeline a regular industry procedure, and that could help to maintain the integrity and optimum efficiency of the pi-
Pigging operations peline and safeguard both the environment and the assets of the pipeline owner. However, unreasonable controls
Fluid-structure interaction of pigging parameters including pressure, pig speed, period, etc. may induce harmful vibrations or even struc-
Vibration response
tural damage, especially for the aerial crossing segment. In this paper, a dynamic model capable of describing
Bifurcations
flow characteristics of accumulations driven by pigs and fluid-structure interactions involving fluid pressure,
centrifugal force and Coriolis force is established to investigate dynamic behaviors of an aerial crossing pipe
during pigging operations. The motion equation is solved using finite element method. Experimental results are
used to validate the motion equation of pipes conveying two-phase flow. A lot of numerical simulations of
dynamic responses under different boundary conditions and operating parameters are performed, and some
interesting and sometimes unexpected results are shown. Finally, the effects of boundary conditions and oper-
ating parameters on dynamic characteristics are discussed. This study is helpful for controlling harmful vibration
responses of natural gas pipeline under pigging operations and promoting the safety of pigging procedures.

1. Introduction
Natural gas transmission pipelines are vital energy artery for in-
dustrial development and the people's livelihood. In order to maintain
the product quality and the integrity and optimum efficiency of the
transportation system and safeguard both the environment and the
assets of the pipeline owner, pigging operations are regularly im-
plemented for long-distance gas pipeline to clean any deposits inside,
such as condensate, water, rust or other solid particles that are often
produced with the natural gas [1].
The high pressure gas from upstream drives pigs to push deposits
accumulating to form a liquid/solid plug during pigging operations.
The deposits flowing through the pipeline can be considered as a two-
phase flow phenomenon [2]. If the operating conditions involving plug
length which is related to yields of deposits and pigging periods, fluid
density depending on deposit species, pig speed and operating pressure
are not controlled reasonably, the pigging operations would cause the
risk of piping instability even damage [3], especially for the aerial
crossing pipelines which are constructed when the pipeline route spans
rivers, roads, railways and other obstacles.
The instability phenomenon of pipes conveying fluid have received
considerable attention and hundreds of relative studies are reported
over the past 40 years, and the instability behaviors of pipes subjected
to single-phase flows is now reasonably well understood. However, the
dynamic behaviors of pipes conveying two-phase flow are difficult to
predict due to complex characteristics of gas-liquid two-phase flow
[4,5] and fluid-structure interactions [6–8]. A number of studies were
performed to investigate the vibrations of tube bundles in two-phase
cross-flow [9–13], and these work were largely related to nuclear steam
generators. Some studies on the vibration behavior of flexible cylinders
subjected to external axial two-phase flow were undertaken from 1970s
for the development of nuclear fuels for boiling water reactors (BWR)
[14–16]. Particularly, a few theoretical models were developed to in-
vestigate internal multi-phase flow induced vibration problems. How-
ever, these models were established under certain conditions and ig-
nored some important factors (operating pressure, transitional velocity,

∗
Corresponding author.
E-mail address: lincw_wang@qq.com (L. Wang).

https://doi.org/10.1016/j.ijpvp.2018.07.011
Received 9 April 2018; Received in revised form 15 July 2018; Accepted 24 July 2018
Available online 25 July 2018
0308-0161/ © 2018 Elsevier Ltd. All rights reserved.
L. Wang et al. International Journal of Pressure Vessels and Piping 165 (2018) 286–294

Fig. 1. Pigging process in modeling.

Le Le
etc.), thus not available for analyzing the dynamic characteristics of
aerial-crossing pipeline under pigging operations [17–23]. Meanwhile, Ce = ∫ N TCSe Ndx + 2∫ N′T [MGe (x, t ) VGe (x, t )
0 0
a better understanding of dynamic behaviors of aerial-crossing pipe-
lines under pigging operations is generally required for pipe structural ⎤
+ MDe (x , t ) VDe (x , t ) ⎥ Ndx
stability and operation safety. (4)
⎦
In this work, a dynamic model capable of describing dynamics be-
haviors of an aerial crossing gas pipeline under pigging operations is Le Le

established, which includes a dynamic model for the fluid conveying Ke = ∫ BTEI Bdx + ∫ BT [P (x, t ) AF − TN (x, t )] Ndx
pipes and pigging model for flow characteristics during pigging op- 0 0
Le
erations. The dynamic model is adopted for the aerial crossing pipe T 2 2

subjected to accumulation plug, and the finite element method is em-

+ ∫ N ′ [MGe (x, t ) VGe (x, t ) + MDe (x, t ) VDe (x, t ) ] N′dx
0 (5)
ployed to solve the corresponding dynamic equations. The experiment
of dynamic response of a horizontal pipe subjected to slug flow is Le

performed to validate the dynamic model and solution methods in this Pe = ∫ [MGe (x, t ) + MDe (x, t ) + m] N Tdx
paper. And then the effects of operating conditions of pigging proce- 0 (6)
dures on dynamic characteristics are evaluated and analyzed. where N and B denote shape function and strain matrices. MGe , VGe , MDe
and VDe are functions in the local coordinate for mass-per-unit-length
and flow velocities of gas and liquid.
2. Modeling approach
2.2. Pigging model
2.1. Motion equation of pipes conveying two-phase flow
For describing major characteristics of pigging phenomenon as
Considering that the different phases have different densities and
shown in Fig. 1, plug zone driven by a pig is considered as a moving
flow velocities, each of the phases is considered to create its own
control volume, and its volume and velocity are almost constant in
Coriolis and centrifugal forces. In addition, the fluid pressure P (x , t ) ,
special pipe section. Ignoring the quality of the pig, the force balance on
axial tension of pipe TN (x , t ) , mass-per-unit-length of gas and dense
the plug formed of liquid/solid accumulations can be expressed as
phase MG (x , t ) and MD (x , t ) , gas and dense phase velocities VG (x , t ) and
VD (x , t ) , lateral displacement u (x , t ) are varying with time and position dVpig d 2
m plug = π ⎛ ⎞ ΔP − m plug g sinθ − Fd
(t is time and x is the coordinate along the centerline of the pipe). The dt ⎝2⎠ (7)
motion equation [24] for a pipe conveying two-phase flow is expressed
as follows: where Fd is drag force of the plug, g is gravitational acceleration,
ΔP = Pup − Pdown is differential pressure between upstream and down-
∂ 4u ∂ ⎡ ∂u stream of the plug, Pup and Pdown are the pressure of upstream and
EI + (P (x , t ) AF − TN (x , t )) ⎤ downstream.
∂x 4 ∂x ⎣ ∂x ⎦
∂ 2u The mass of the accumulation plug m plug is
+ [MG (x , t ) VG2 (x , t ) + MD (x , t ) VD2 (x , t )]
∂x 2 d 2

∂ 2u m plug = π ⎛ ⎞ ρplug L plug

+ 2[MG (x , t ) VG (x , t ) + MD (x , t ) VD (x , t )] ⎝2⎠ (8)
∂x ∂t
∂ 2u ∂u where d is inner diameter of the pipe, ρplug is the density of the accu-
+ [MG (x , t ) + MD (x , t ) + m] 2 + Cs mulation plug. L plug is the plug length.
∂t ∂t
The pig speed Vpig is equal to the fluid velocity in the plug [25], and
+ [MG (x , t ) + MD (x , t ) + m] g = 0 (1)
can be expressed as
The terms of Eq. (1) represents respectively: flexural force, effect of 4P0 T Z Q In
Vpig =
pressure and axial force, centrifugal force, Coriolis force, inertia force, πd 2T0 Z0 P (9)
structural damping force and gravity. The effect of pressure and axial
force, centrifugal force and Coriolis force are fluid-structure interac- where P0 = 101.325 kPa, T0 = 293.15 K and Z0 = 1. Z is the average
tions, which affect the dynamic characteristics of the FSI system. Finite compressibility factor of upstream gas.
element method is used to solve the motion equations, and the dynamic The average pressure P and the average temperature T can be ex-
equation for each element is expressed in matrix form as, pressed as [25].
2
¨ e + Ce U˙ e + K e Ue = Pe 2⎛ Pup ⎞
Me U (2) P = PIn +
3 ⎜⎝ PIn + Pup ⎟⎠ (10)
where Me, Ce and Ke denote mass, damping and stiffness matrices, and
Pe is the element nodal force vector, and can be calculated by, 1 − e−aL
T = Te + (TIn − Te)
aL (11)
Le
Me = ∫ NT [MGe (x , t) + MDe (x , t ) + m] Ndx
a=
k πD
0 (3) mg Cp (12)

287
L. Wang et al. International Journal of Pressure Vessels and Piping 165 (2018) 286–294

where PIn is the pressure at starting inlet, Te is environment tempera- Table 1

ture, k is total heat transfer coefficient, D is outer diameter, mg is gas Measured slug parameters in selected time domain.
mass flow, Cp is specific heat at constant pressure. Slug No. LS (m) VT (m/s) tEN (s)

2.3. Coupling computation of fluid and structure 1 1.19 3.50 1.50

2 1.09 3.03 4.85
3 0.73 3.64 8.20
The two-phase flow parameters in Eq. (1) can be calculated as fol- 4 1.19 2.33 11.30
lowing.
For plug zone
MD = ρplug AF superficial gas velocity VSG is 1.39 m/s, and the superficial liquid ve-
(13)
locity VSL is 0.29 m/s. Then, the pressure signals and displacement
MG = 0 (14) signals are measured.
Three pressure sensors are paired to form two differential pressure
VD = Vpig (15) (DP) transducers. The two DP signals of one liquid slug could be treated
For other zone using cross-correlation algorithm [31] to obtain the transitional velo-
city and the length of the liquid slug. Four liquid slugs flow in the
MD = 0 (16) testing segment during the selected time domain, and the measured
MG = ρG AF (17) parameters of the liquid slugs are presented in Table 1. The reduced
liquid height of liquid film zone is measured as 0.33, and the liquid
VG = Vpig (18) holdup of slug body is estimated by Essama's correlation [32]:
PR EoS [26] is considered as the accurate equation of state to HLS = (1.009 − CM VM) A (25)
compute the thermophysical properties of natural gas [27–29]. There-
1, θ ≤ 0
where CM = 0.006 + 1.3377 μG , VM = VSL + VSG , A = ⎧
μ
fore, the gas density ρG (equal to 1/ v ) is estimated using PR EoS: .
⎩1 − sinθ , θ > 0
⎨
L
RT a (T ) The dynamic response of the pipe midpoint is simulated by present
P= −
v−b v (v + b) + b (v − b) (19) model. The initial posture of pipes is set as the static equilibrium. The
dynamic response of horizontal pipe subjected to hydrodynamic slug
where P is pressure, R is gas constant, v is specific volume, a and b are
flow is presented in Fig. 3, and the theory results are validated by ex-
the constants relating to the gas compositions.
perimental data. The details of vibration response after slug leaving are
a (T ) = ac α(T ) (20) described in Fig. 3b. The numerical results are matched with the ex-
perimental results, including the response amplitude and period. It can
R2Tc2
ac = 0.45724 be concluded that the dynamic response is relative to slug character-
Pc (21) istics involving slug length, period and transitional velocity. The vi-
RTc bration response occurs after the liquid slug leaving due to the sudden
bc = b = 0.07780 change of dynamics characteristics (including stiffness, mass, damping
Pc (22)
and loading) of the FSI system during the liquid slug transition.
α (T ) = [1 + K (1 − Tr0.5)]2 (23)
4. Simulations and discussions
K = 0.37464 + 1.54226ω − 0.26992ω2 (24)
where Tr is reduced temperature, K is empirical coefficient, ω is acentric In order to understand the complicated dynamic responses and in-
factor, Tc and Pc are the critical temperature and pressure. stability of the aerial-crossing gas pipelines under pigging operations,
the dynamic response and stability analysis of an aerial-crossing girder
3. Model validation pipeline spanning a river in Shengli oilfield are performed as a case
study. The outer diameter is 323.9 mm, the wall thickness is 7.1 mm
Validation experiments were carried out on an air-water pipeline and the span length is 16.0 m. The constraints at two ends are con-
loop to measure the dynamic displacement of the horizontal pipe sidered as Fix-Fix and Fix-Pin respectively to simulate the dynamic
caused by liquid slug in order to validate the theoretical model of pipes responses of the aerial-crossing pipe under different boundary condi-
conveying two-phase flow. Parameters of the testing segment and in- tions. In this section, the dynamic behaviors of the aerial-crossing pipe
struments, schematic process and scene graph of experimental appa- during pigging operations are simulated in various operating para-
ratus are described detailedly in Ref. [24]. The pressure sensors (P1, P2 meters and boundary conditions, and the effects of fluid density, pig
and P3) and the displacement sensors (D1, D2 and D3) were installed on speed, plug length and operating pressure on dynamic behaviors are
the testing segment to collect pressure signals and displacement signals, analyzed.
and the installation locations are described in Fig. 2.
According to the flow pattern map in Ref. [30], the flow rates of air 4.1. Effects of fluid density on dynamic response
and water are determined and verifications are performed to make sure
that hydrodynamic slugs occur stably in the testing segment. The According to field practices, the accumulations removed from gas-

Fig. 2. Schematic diagram of the measure point on

testing segment.

288
L. Wang et al. International Journal of Pressure Vessels and Piping 165 (2018) 286–294

Fig. 3. Comparison of simulation and experimental results of dynamic response: (a) measure point D1, (b) measure point D2 and (c) measure point D3
(VSG = 1.39 m/s, VSL = 0.29 m/s).

transmission pipeline involve condensate, water, rust, foam debris and
other solid particles. The types of accumulations depend on gas sources,
pre-treatment process and operating conditions. To analyze dynamic
behaviors of the aerial-crossing pipe with different deposits, dynamic
behaviors with different fluid densities are simulated.
Fig. 4 presents dynamic responses of the midpoint on the pipe under
pigging operations, and the important characteristics including the
static equilibrium position (Dequ), maximum load deflection (maximum
displacement during the plug passing, Dmax) and maximum vibration
amplitude (maximum vibration amplitude after the plug leaves, Amax)
are marked. Fig. 5 and Fig. 6 present bifurcation diagrams for three
positions. In the bifurcation diagrams, the data points represent the
extreme values of vibration displacements under different parameters,
and that can describe the vibration characteristics under different
conditions. As shown in Figs. 5 and 6, Dmax increases gradually with
fluid density increasing, but the Amax increase intensely after the fluid
density reaches 3000 kg/m3 and 1500 kg/m3 when the boundary con-
ditions are Fix-Fix and Fix-Pin respectively. It indicates that higher fluid
density is easier to induce the divergence of vibration amplitudes even
when fluid velocity is small, because the gravitational load, centrifugal
force and Coriolis force are increasing with fluid density. Therefore, the
controls of the accumulation types and heavy component content in the
pipe are essential for pigging operations.

Fig. 4. Dynamic response of the midpoint with different fluid densities, Vpig = 5.5 m/s, Lplug = 6.7 m and Pop = 3 MPa, boundary condition: (a) Fix-Fix; (b) Fix-Pin.

289
L. Wang et al. International Journal of Pressure Vessels and Piping 165 (2018) 286–294

Fig. 5. Bifurcation diagrams for three positions of a system defined by Vpig = 5.5 m/s, Lplug = 6.7 m and Pop = 3 MPa as ρplug is varied, constraint condition: Fix-Fix,
(a) 1/4; (b) 1/2; (c) 3/4.

Fig. 6. Bifurcation diagrams for three positions of a system defined by Vpig = 5.5 m/s, Lplug = 6.7 m and Pop = 3 MPa as ρplug is varied, constraint condition: Fix-Pin,
(a) 1/4; (b) 1/2; (c) 3/4.

Fig. 7. Dynamic response of the midpoint with different pig speeds, ρplug = 1000 kg/m3, Lplug = 6.7 m and Pop = 3 MPa, constraint condition: (a) Fix-Fix; (b) Fix-Pin.

4.2. Effects of pig speed on dynamic response 4.3. Effects of plug length on dynamic response

The effects of pig speed on the dynamic characteristics of aerial
crossing pipe are evaluated here. Fig. 7 presents time-domain responses
of the pipe midpoint with different pig speeds. With the pig speed in-
creasing, the duration of plug flowing in aerial crossing pipe becomes
shorter and the free-vibration amplitude after plug leaving varies. The
important characteristic parameters of dynamic behaviors under dif-
ferent pig speeds can be drawn in Fig. 8 and Fig. 9, which present the
extreme values of the displacements during the dynamic responses. It is
observed that Amax increases intensely when the pig speed is higher
than 20 m/s and 15 m/s under two boundary conditions respectively.
Amax would vary when the pig speed is increasing, and Amax would
diverge under different boundary conditions if the pig speed is larger
than 300 m/s and 250 m/s respectively. That can be explained that the
fluid-structure interactions including the centrifugal force and Coriolis
force are increasing but the response time decreases with pig speed. It
indicates the control of pig speed is significantly crucial for the safety of
pigging operations.
The plug length reflects the accumulation volume in the pipe.
Effects of plug length on the dynamic characteristics of aerial crossing
pipe are evaluated here. Fig. 10 presents time-domain responses of the
pipe midpoint with different plug lengths. It is observed that the
maximum load deflection increases with the plug length increasing till
the plug length reaches pipe length. Fig. 11 and Fig. 12 present the
extreme values of the displacements during the dynamic responses with
different plug lengths. Dmax increases with the plug length when the
plug length is less than the pipe length, and keeps constant when the
plug length is larger than the pipe length. Amax would be constant when
the plug length is respectively larger than 15 m and 30 m under dif-
ferent boundary conditions. It indicates that the effect of the accumu-
lation volume on the vibration amplitude is limited.
4.4. Effects of operating pressure on dynamic response
Operating pressure (the upstream pressure of the pig) is significant
controlled parameter in pigging operations and affects the dynamic

290
L. Wang et al.
Fig. 8. Bifurcation diagrams for three positions of a system defined by
ρplug = 1000 kg/m3, Lplug = 6.7 m and Pop = 3 MPa as Vpig is varied, constraint
condition: Fix-Fix, (a) 1/4; (b) 1/2; (c) 3/4.
behaviors of gas pipeline. We evaluate the effects of operating pressure
on the dynamic characteristics of aerial crossing pipe neglecting the
friction drag of the plug. Fig. 13 shows time-domain dynamic responses
of aerial crossing pipe under different operating pressures. It is
291
International Journal of Pressure Vessels and Piping 165 (2018) 286–294
Fig. 9. Bifurcation diagrams for three positions of a system defined by
ρplug = 1000 kg/m3, Lplug = 6.7 m and Pop = 3 MPa as Vpig is varied, constraint
condition: Fix-Pin, (a) 1/4; (b) 1/2; (c) 3/4.
observed that the deflection and vibration amplitude are larger for the
pipes in higher operating pressure. It indicates that higher internal
pressure can reduce the stiffness of the pipe. Bifurcation diagrams of
dynamic responses under different operating pressures are shown in
Fig. 14 and Fig. 15. Dequ and Dmax increase with the operating pressure
increasing. Amax would diverge when the operating pressure is
L. Wang et al. International Journal of Pressure Vessels and Piping 165 (2018) 286–294

Fig. 10. Dynamic response of the midpoint with different plug lengths, ρplug = 1000 kg/m3, Vpig = 5.5 m/s and Pop = 3 MPa, constraint condition: (a) Fix-Fix; (b) Fix-
Pin.

Fig. 11. Bifurcation diagrams for three positions of a system defined by ρplug = 1000 kg/m3, Vpig = 5.5 m/s and Pop = 3 MPa as Lplug is varied, constraint condition:
Fix-Fix, (a) 1/4; (b) 1/2; (c) 3/4.

Fig. 12. Bifurcation diagrams for three positions of a system defined by ρplug = 1000 kg/m3, Vpig = 5.5 m/s and Pop = 3 MPa as Lplug is varied, constraint condition:
Fix-Pin, (a) 1/4; (b) 1/2; (c) 3/4.

Fig. 13. Dynamic response of the midpoint with different operating pressure, ρplug = 1000 kg/m3, Vpig = 5.5 m/s and Lplug = 6.7 m, constraint condition: (a) Fix-Fix;
(b) Fix-Pin.

292
L. Wang et al. International Journal of Pressure Vessels and Piping 165 (2018) 286–294

Fig. 14. Bifurcation diagrams for three positions of a system defined by ρplug = 1000 kg/m3, Vpig = 5.5 m/s and Lplug = 6.7 m as Pop is varied, constraint condition:
Fix-Fix, (a) 1/4; (b) 1/2; (c) 3/4.

Fig. 15. Bifurcation diagrams for three positions of a system defined by ρplug = 1000 kg/m3, Vpig = 5.5 m/s and Lplug = 6.7 m as Pop is varied, constraint condition:
Fix-Pin, (a) 1/4; (b) 1/2; (c) 3/4.

respectively larger than 30 MPa and 15 MPa under different boundary express their gratitude to the support from the National Natural Science
conditions. It indicates that high operating pressure could induce the Fund (No. 11772273) and the National Natural Science Fund for
instability of the pipe during pigging operations. Distinguished Young Scholars (No. 51404290).

5. Conclusions Appendix A. Supplementary data

The dynamic behaviors and bifurcation characteristics of an aerial Supplementary data related to this article can be found at https://
crossing pipeline under pigging operations were investigated numeri- doi.org/10.1016/j.ijpvp.2018.07.011.
cally. The present theoretical model was extended to consider two-
phase flow characteristics and fluid-structure interactions, and vali- Nomenclature
dated with experimental results. The complicated dynamic behaviors
under different conditions were simulated, and the characteristics were AF Area of the flow channel cross section
analyzed. The typical conclusions are summarized as follows: Amax Maximum vibration amplitude after plug leaves
Cp Specific heat at constant pressure
(1) The dynamic responses of the system are the cumulative effects of Cm Mixture velocity coefficient
fluid-structure interactions and two-phase flow characteristics Cs Damping coefficient
during pigging operations. D, d Outer and inner diameters of pipe
(2) The fluid density and pig speed affect the centrifugal force, Coriolis Dequ, Dmax Equilibrium position and max load deflection
force and the rate of change of system properties, thus have sig- E, I Elastic modulus and moment of inertia
nificant influence on the maximum deflection and can excite in- Fd Drag force
tense vibration response after plug leaving. Boundary conditions g Gravitational acceleration
have significant influence on vibration characteristics. HLS Liquid holdup of liquid slug
(3) The effects of the accumulation volume on the vibration responses k Total heat transfer coefficient
are limited. However, higher levels of pig speed, fluid density or L plug Length of the plug
operating pressure are easier to induce large vibration amplitudes, Le Length of pipe element
and these parameters are crucial for the stability of the pipeline and m plug Mass of the plug
should be controlled reasonably during pigging operations. mg Gas mass flow
(4) The results can enrich our understanding on complicated dynamics MG , MD, m Gas, dense phase and pipe mass-per-unit-length
behaviors of natural gas pipelines during pigging operations and MGe , VGe , MDe VDe Gas, dense phase mass-per-unit-length and flow velo-
may promote the operation safety of natural gas pipelines. cities in local coordinate
P Local pressure of the fluid
Pc , Tc Critical pressure and temperature
Acknowledgements
P , T , Z Average pressure, temperature and compressibility factor in
pipeline
For the work described in this paper, the authors would like to

293
L. Wang et al. International Journal of Pressure Vessels and Piping 165 (2018) 286–294

Pdown , Pup Pressure at downstream and upstream 1.2929583.

PIn Pressure at starting inlet and plug zone [12] M.J. Pettigrew, C.E. Taylor, N.J. Fisher, M. Yetisir, B.A.W. Smith, Flow-induced
vibration: recent findings and open questions, Nucl. Eng. Des. 185 (1998) 249–276,
Pop Operating pressure https://doi.org/10.1016/S0029-5493(98)00238-6.
Tc Oscillating period of coupling vibration [13] J. Choi, E. Pereyra, C. Sarica, C. Park, J. Kang, An efficient drift-flux closure re-
Te Environment temperature lationship to estimate liquid holdups of gas-liquid two-phase flow in pipes, Energies
5 (2012) 5294–5306, https://doi.org/10.3390/en5125294.
TN Axial tension of pipe [14] M.P. Paidoussis, M.J. Pettigrew, Dynamics of flexible cylinders in axisymmetrically
Tr Reduced temperature confined axial flow, J. Appl. Mech. 46 (1979) 37–44, https://doi.org/10.1115/1.
t Time 3424525.
[15] J.L. Riverin, E. de Langre, M.J. Pettigrew, Fluctuating forces caused by internal two-
tEN Entrance time phase flow on bends and tees, J. Sound Vib. 298 (2006) 1088–1098, https://doi.
u Lateral displacement org/10.1016/j.jsv.2006.06.039.
v Specific volume [16] B.L. Tay, R.B. Thorpe, Hydrodynamic forces acting on pipe bends in gas-liquid slug
flow, Chem. Eng. Res. Des. 92 (2014) 812–825, https://doi.org/10.1016/j.cherd.
VG , VD Flow velocities of gas and dense phase
2013.08.012.
VM , CM Mixture velocity, mixture velocity coefficient [17] C. Monette, M.J. Pettigrew, Fluidelastic instability of flexible tubes subjected to
Vpig Pig speed two-phase internal flow, J. Fluid Struct. 19 (2004) 943–956, https://doi.org/10.
VSG , VSL Local superficial velocity of gas and liquid 1016/j.jfluidstructs.2004.06.003.
[18] L. Chica, FSI Study of Internal Multiphase Flow in Subsea Piping Components,
x Position along the axis (2014).
Me , Ce, K e Mass, damping, stiffness matrices [19] A. Ortega, A. Rivera, C.M. Larsen, Flexible riser response induced by combined slug
N, B Shape function and strain matrices flow and wave loads, ASME 2013 32nd Int. Conf. Ocean. Offshore Arct. Eng, 2013,
p. 10, , https://doi.org/10.1115/OMAE2013-10891.
Ue, Pe Displacement vector and nodal force vector of pipe element [20] F. Hara, Two-phase-flow-induced vibrations in a horizontal piping system, Bull.
θ Inclination angle of the pipe from horizontal JSME 20 (1977) 419–427.
ρG , ρplug Densities of gas, gas core and liquid [21] X. Xu, M. Liu, Y. Ma, M. An, Effects of fluidized solid particles on vibration beha-
viors of a graphite tube evaporator with an internal vapor-liquid flow, Appl. Therm.
μG , μL Dynamic viscosity of gas and liquid Eng. 100 (2016) 1229–1244, https://doi.org/10.1016/j.applthermaleng.2015.12.
ω Acentric factor 126.
[22] L. Wang, Y. Yang, C. Liu, Y. Li, Q. Hu, Numerical investigation of dynamic response
of a pipeline-riser system caused by severe slugging flow, Int. J. Pres. Ves. Pip. 159
References (2018) 15–27, https://doi.org/10.1016/j.ijpvp.2017.11.002.
[23] L. Wang, Y. Yang, Y. Li, Y. Wang, Resonance analyses of a pipeline-riser system
[1] Y. Li, S. Feng, Simulation of pigging dynamics in two-phase flow pipelines, J. Chem. conveying gas-liquid two-phase flow with flow-pattern evolution, Int. J. Pres. Ves.
Ind. Eng. 55 (2004) 271–274. Pip. 161 (2018) 22–32, https://doi.org/10.1016/j.ijpvp.2018.01.005.
[2] B. Jamshidi, M. Sarkari, Simulation of pigging dynamics in gas-liquid two-phase [24] L. Wang, Y. Yang, Y. Li, Y. Wang, Dynamic behaviours of horizontal gas-liquid pipes
flow pipelines, J. Nat. Gas Sci. Eng. 32 (2016) 407–414, https://doi.org/10.1016/j. subjected to hydrodynamic slug flow: modelling and experiments, Int. J. Pres. Ves.
jngse.2016.04.018. Pip. 161 (2018) 50–57, https://doi.org/10.1016/j.ijpvp.2018.02.005.
[3] J. Cordell, H. Vanzant, The Pipeline Pigging Handbook, (2003). [25] Y. Li, G. Yao, The Design and Management of Gas Transmission Pipeline, Second,
[4] G. Xu, L. Cai, A. Ullmann, N. Brauner, Experiments and simulation of water dis- UPC Press, Dongying, 2009.
placement from lower sections of oil pipelines, J. Petrol. Sci. Eng. 147 (2016) [26] D. Peng, D.B. Robinson, A new two-constant equation of state, Ind. Eng. Chem.
829–842, https://doi.org/10.1016/j.petrol.2016.09.049. Fundam. 15 (1976) 59–64, https://doi.org/10.1021/i160057a011.
[5] Y. Yang, J. Li, S. Wang, C. Wen, Gas-liquid two-phase flow behavior in terrain- [27] Y. Yang, C. Wen, CFD modeling of particle behavior in supersonic flows with strong
inclined pipelines for gathering transport system of wet natural gas, Int. J. Pres. Ves. swirls for gas separation, Separ. Purif. Technol. 174 (2017) 22–28, https://doi.org/
Pip. 162 (2018) 52–58, https://doi.org/10.1016/j.ijpvp.2018.03.005. 10.1016/j.seppur.2016.10.002.
[6] M.P. Paidoussis, Fluid-structure Interactions: Slender Structures and Axial Flow, [28] Y. Yang, J.H. Walther, Y. Yan, C. Wen, CFD modeling of condensation process of
second ed., (2014), https://doi.org/10.1017/CBO9781107415324.004. water vapor in supersonic flows, Appl. Therm. Eng. 115 (2017) 1357–1362, https://
[7] S.S. Chen, Flow-induced Vibration of Circular Cylindrical Structures, (1985), doi.org/10.1016/j.applthermaleng.2017.01.047.
https://doi.org/10.2172/6331788. [29] L. Teng, Y. Li, Q. Zhao, W. Wang, Q. Hu, X. Ye, D. Zhang, Decompression char-
[8] M.P. Païdoussis, Fluid-structure Interactions: Slender Structures and Axial Flow, acteristics of CO2 pipelines following rupture, J. Nat. Gas Sci. Eng. 36 (2016)
second ed., (2016), https://doi.org/10.1016/C2011-0-08058-4. 213–223, https://doi.org/10.1016/j.jngse.2016.10.026.
[9] F. Baj, E. de Langre, Scaling of damping induced by bubbly flow across tubes, J. [30] H.-Q. Zhang, Q. Wang, C. Sarica, J.P. Brill, Unified model for gas-liquid pipe flow
Fluid Struct. 17 (2003) 351–364, https://doi.org/10.1016/S0889-9746(02) via slug Dynamics—Part 2: model validation, J. Energy Resour. Technol. 125
00142-1. (2003) 274, https://doi.org/10.1115/1.1615618.
[10] I.C. Chu, H.J. Chung, S. Lee, Flow-induced vibration of nuclear steam generator U- [31] L. Xu, The state-of-art and developing trend of correlated flow measuring techni-
tubes in two-phase flow, Nucl. Eng. Des. 241 (2011) 1508–1515, https://doi.org/ ques, Process Autom. Instrumentation. 9 (1988) 1–4.
10.1016/j.nucengdes.2011.01.034. [32] C. Omgba-Essama, Numerical Modelling of Transient Gas-liquid Flows (Application
[11] M.J. Pettigrew, C.E. Taylor, Two-phase flow-induced vibration: an overview, J. to Stratified & Slug Flow Regimes), Cranfield University, 2004.
Press. Vessel Technol. Trans. ASME 116 (1994) 233–253, https://doi.org/10.1115/

294

View publication stats