processes

Article
Solid–Liquid Two-Phase Flowmeter Flow-Passage Wall Erosion
Evolution Characteristics and Calibration of
Measurement Accuracy
Wei Han 1,2 , Lumin Yan 1 , Rennian Li 1,3, *, Jing Zhang 1 , Xiang Yang 1 , Lei Ji 4 and Yan Qiang 1

1 School of Energy and Power Engineering, Lanzhou University of Technology, Lanzhou 730050, China
2 Key Laboratory of Advanced Pumps Valves and Fluid Control System of the Ministry of Education,
Lanzhou University of Technology, Lanzhou 730050, China
3 Key Laboratory of Fluid Machinery and Systems, Lanzhou 730050, China
4 Key Laboratory of Fluid Machinery and Engineering, Xihua University, Chengdu 610039, China
* Correspondence: 212080704026@lut.edu.cn

Abstract: Solid–liquid two-phase flowmeters are widely used in critical sectors, such as petrochem-
icals, energy, manufacturing, the environment, and various other fields. They are indispensable
devices for measuring flow. Currently, research has primarily focused on gas–liquid two-phase flow
within the flowmeter, giving limited attention to the impact of solid phases. In practical applications,
crude oil frequently contains solid particles and other impurities, leading to equipment deformation
and a subsequent reduction in measuring accuracy. This paper investigates how particle dynamic
parameters affect the erosion evolution characteristics of flowmeters operating in solid–liquid two-
phase conditions, employing the dynamic boundary erosion prediction method. The results indicate
that the erosion range and peak erosion position on the overcurrent wall of the solid–liquid two-phase
flowmeter vary with different particle dynamic parameters. Erosion mainly occurs at the contraction
section of the solid–liquid two-phase flowmeter. When the particle inflow velocity increases, the
erosion range shows no significant change, but the peak erosion position shifts to the right, primarily
due to the evolution of the erosion process. With an increase in particle diameter, the erosion range
expands along the inlet direction due to turbulent diffusion, as particles with lower kinetic energy
Citation: Han, W.; Yan, L.; Li, R.;
exhibit better followability. There is no significant change in the erosion range and peak erosion
Zhang, J.; Yang, X.; Ji, L.; Qiang, Y.
position with an increase in particle volume fraction and particle sphericity. With a particle inflow
Solid–Liquid Two-Phase Flowmeter
velocity of 8.4 m/s, the maximum erosion depth reaches 750 µm. In contrast, at a particle sphericity
Flow-Passage Wall Erosion Evolution
of 0.58, the minimum erosion depth is 251 µm. Furthermore, a particle volume fraction of 0.5 results
Characteristics and Calibration of
Measurement Accuracy. Processes
in a maximum flow coefficient increase of 1.99 × 10−3 .
2024, 12, 1012. https://doi.org/
10.3390/pr12051012 Keywords: particle dynamics parameters; erosion characteristics; measurement accuracy

Academic Editor: Blaž Likozar

Received: 9 April 2024

Revised: 10 May 2024 1. Introduction
Accepted: 14 May 2024 In the current energy market, oil plays a crucial role. With the rapid global advance-
Published: 16 May 2024 ment of science, technology, and the economy, the demand for oil has escalated. The
consequential surge in transactions, coupled with the depletion of onshore oil reserves, has
become an inescapable reality. Exploiting submarine fields emerges as a developmental
trend in the oil and gas sector to alleviate the growing disparity between supply and
Copyright: © 2024 by the authors.
demand [1]. In the extraction, transport, and trading of sub-sea oil fields, the accuracy and
Licensee MDPI, Basel, Switzerland.
This article is an open access article
performance of the solid–liquid two-phase flowmeter play a crucial role for various stake-
distributed under the terms and
holders [2–6]. However, deep-sea crude oil is a complex mixture containing natural gas,
conditions of the Creative Commons solid particles, impurities, and oil. Therefore, it is essential to ensure the measurement ac-
Attribution (CC BY) license (https:// curacy and performance of the solid–liquid two-phase flowmeter. The impact and abrasion
creativecommons.org/licenses/by/ of heavy-phase particles during the solid mixing and transport process in the flowmeter
4.0/). measurement channel cause the deformation of the flow coefficient [7–11]. Consequently,

Processes 2024, 12, 1012. https://doi.org/10.3390/pr12051012 https://www.mdpi.com/journal/processes

Processes 2024, 12, 1012 2 of 13

this decline in measurement performance and accuracy can lead to disputes and losses
in the process of crude oil extraction and transport. Therefore, additional research on the
erosion characteristics of solid–liquid two-phase flowmeters holds significant engineering
importance [12–15]. It aims to safeguard solid–liquid two-phase flowmeters from erosion
and enhance measurement accuracy in practical engineering applications.
Scholars from domestic and international institutions have researched erosion damage
to the flow-passage walls in multiphase flowmeters with fixed boundaries. These factors
encompass the material and shape of the flow-passage walls, particle properties, particle
impact velocity, particle impact angle, and fluid properties [16–21]. Their interactions
collectively contribute to the individual impact on the equipment. Li [22] designed an
experimental setup to measure the impact of single particles and explored the influence of
impact parameters on particle behaviour and material deformation. Their findings revealed
that impact velocity significantly influences particle motion parameters, yet the shape of
the erosion crater remains consistent across various impact velocities. Moreover, He [12]
proposed a correlation equation for measuring the wet gas flow of the venturi flowmeter
based on the two-phase mass flow coefficient. They discussed the factors influencing
this coefficient and observed a linear increase with the Lockhart–Martinelli parameter,
along with a decrease as the gas–liquid density ratio increased. Furthermore, various
factors influence the erosion and damage to equipment operating in sandy conditions.
Gajan [23] conducted high-pressure experiments on a venturi flowmeter in wet gas flow
conditions. They analyzed the impact of water content on flow behaviour, comparing
experimentally obtained flow coefficients with predictions derived from the flowmeter’s
internal dynamics. Similarly, Dehkordi [24] employed the volume of fluid (VOF) model
to investigate the two-phase flow of high-viscosity oil and water in a venturi flowmeter.
They acquired data on two-phase pressure drops, instantaneous radial velocity, holding
rate distributions, cross-section time-averaged holding rates, and slip rates. Lastly, Liu [25]
conducted numerical calculations on a low-temperature venturi flowmeter. They observed
a linear relationship between the flow coefficient and the root mean square of the reciprocal
of the Reynolds number of the throat. The flow coefficient decreased with an increase in
the diameter ratio of the throat and increased with a rise in the angle of constriction.
Currently, the primary focus in multiphase flowmeter research revolves around inter-
nal flow characteristics and cavitation. However, erosion factors have not been adequately
considered in these studies. Furthermore, erosion-related research often relies on fixed
boundaries, lacking a genuine representation of flow field conditions. This paper exam-
ines the impact of various particle dynamic parameters on the erosion characteristics and
measurement accuracy of the solid–liquid two-phase flowmeter. It employs the dynamic
boundary erosion prediction method along with the Oka erosion model to offer a theoretical
reference for enhancing the measurement accuracy of solid–liquid two-phase flowmeters.

2. Numerical Model
2.1. Computational Domain Model
To comply with confidentiality requirements, the pertinent structural parameters of the
solid–liquid two-phase flowmeter cannot be disclosed. The computational domain is estab-
lished by SolidWorks 2010 3D modeling software, which includes the inlet straight section,
tapered section, throat, expanding section, and the outlet straight section. Refer to Figure 1
for the profile of the solid–liquid two-phase flowmeter computational domain model.
Processes 2024, 12, x FOR PEER REVIEW 3 of 14

Processes2024,
Processes 2024,12,
12,1012
x FOR PEER REVIEW 3 of 14
3 of 13

Figure 1. Solid–liquid two-phase flowmeter profile.

Figure1.1. Solid–liquid
Figure Solid–liquidtwo-phase
two-phaseflowmeter
flowmeterprofile.
profile.
2.2. Meshing of the Computational Domain
2.2.
The Meshing
2.2.erosion
Meshing of
of the
the Computational
evolution Domain
characteristics
Computational of solid–liquid two-phase flowmeters are stud-
Domain
ied in this The
papererosion
usingevolution
The erosion dynamiccharacteristics
evolution mesh of solid–liquid
technology.
characteristics
two-phase
The unstructured
of solid–liquid
flowmeters
mesh
two-phase
are studied
deformation
flowmeters are stud-
in thisduring
that occurs paper movement
using dynamic
is mesh technology.
reconstructed be er The
to unstructured
enhance the meshmesh deformation
quality. There- that
ied in this paper using dynamic mesh technology. The unstructured mesh deformation
occurs
fore, the during
computationalmovement is reconstructed better to enhance the mesh quality. Therefore, the
that occurs duringdomain model
movement is divided intobeaner
is reconstructed unstructured
to enhance themesh. Structured
mesh quality. There-
meshingcomputational
of the domain
thecomputational model
computationaldomain is divided
domainmodel into an
was performedunstructured
using mesh. Structured meshing of
fore, is divided into an the commerciamesh.
unstructured software
Structured
the
ICEM CFD computational
16.0 To be er domain was performed
capture the near-wall using the commercia software ICEM CFD 16.0
meshing of the computational domain surface flow, covering
was performed using the
theboundary
commercialayer
software
To better
mesh near the capture
wall the near-wall
surface, the surface flow,
computational covering
domain mesh the
is boundaryinlayer
displayed mesh
Figure 2. near the
ICEM CFD 16.0 To be er capture the near-wall surface flow, covering the boundary layer
wall surface, the computational domain mesh is displayed in Figure 2.
mesh near the wall surface, the computational domain mesh is displayed in Figure 2.

(a) (b) (c)

Figure Figure
2. Computational (a)domaindomain
2. Computational model model
meshing.
meshing. (b)
(a) Inlet straight
(a) tapered;
Inlet straight (b) throat;
tapered; (c)(c)
expanding
(b) throat; (c) expanding
outlet straight.
outlet
Figurestraight.
2. Computational domain model meshing. (a) Inlet straight tapered; (b) throat; (c) expanding
outlet straight.
2.3.
2.3. Oka Oka Erosion
Erosion Model Model
The The Erosion
2.3.process
Oka process of particle
of particle
Model collisioncollision
with awith a material
material surface surface
is highly is highly
complex; complex; thus, the
thus, the
erosionerosion
model model
should should
a emptattempt
to to encompass
encompass pertinent pertinent
parameters parametersthat that impact
impact the the degree
degree
The process of particle collision with a material surface is highly complex; thus, the
of erosion.
of erosion. Okamodel Oka [26,27]
[26,27] carried carried anout an experimentresulted that resulted in the in the development of an
erosion should a out
empt experiment
to encompassthat pertinent parameters development
that impactofthe andegree
erosion
erosionofrate rate calculation
calculation formula
formula based based
onan on
Oka’s Oka’s model.
model. that The
The resulted developed
developed formula
formula considered
consid- of an
erosion. Oka [26,27] carried out experiment in the development
variousparameters,
ered various parameters, such such asas particle
particle diameter,
diameter, target
target material
material density,
density, andand
Vickers hardness,
Vickers
erosion rate calculation formula based on Oka’s model. The developed formula consid-
that
hardness, influence the degree of erosion damage. The definition of erosion rate is as follows:
eredthat influence
various the degree
parameters, suchof as
erosion
particle damage.
diameter, The target definition of erosion
material rateand
density, is asVickers
follows:hardness, that influence the degree ofEerosion damage. The definition of erosion rate is(1) as
θ = E90 f ( θ )
follows: E   E 9 0 f   (1)
!k2 !k3
E  V E 9 0 f   d p (1)
E90 = K ( Hv )k1 k2 k3 (2)
V re f d  pre f d
k  V  p
E90  K   v  1  n   k2 k3 (2)

f (θ ) = (sin θ)Vref[1k+
1   HVdv (pref1 −
 sin d pθ )] n2 (3)
E90  K   v     
1
 (2)
The variable θ represents the collision angle;
 V K
 
denotes d 
a constant associated with the
 ref   ref  p
particle properties; Hv denotes f    the
sinVickers
n1
  hardness n2 target material (Gpa); V stands
of the
1   v  1  sin (3)
for the collision velocity (m/s); Vref is the reference n1 velocity (m/s); n2 dp indicates the particle
diameter (m); and dpref f   the
  sin   1   v 1  sin of the k1 , k2 , k(3)
The variable θ represents therepresents
collision angle; reference
K denotes diameter a constant particles with
associated (m). the 3,
n1properties;
particle , and n2 areHconstants
v denotes the related to the
Vickers characteristics
hardness of the targetof the particles
material andVwall
(Gpa); material,
stands
The variable θ represents the collision angle; K denotes a constant associated with the
for thethe valuesvelocity
collision of the parameters
(m/s); Vref is are
the K = 65, dpref = 0.326 mm,dpVindicates
ref = 104 the m/s, k1 = −0.12,
particle properties;
0.038
Hv denotes the reference
Vickers
0.14
velocity
hardness (m/s);
of the
− 0.94
target particle
material (Gpa); V stands
kfor
2 = the
2.3H v
collision , k3velocity
= 0.19, n(m/s);
1 = 0.71HVref vis the , and n2 = 2.4H
reference v
velocity .(m/s); dp indicates the particle
 and n2 are constants related to the characteristics of the particles and wall material, the
values of the parameters are K = 65, dpref = 0.326 mm, Vref = 104 m/s, k1 = −0.12, k2 = 2.3Hv0.038,
k3 = 0.19, n1 = 0.71Hv0.14, and n2 = 2.4Hv−0.94.

Processes 2024, 12, 1012 2.4. Erosion Depth Model Based on Dynamic Boundary 4 of 13

Material surface damage induced by erosion is a nonlinear, time-varying process

with spatial characteristics. The parameters of particle dynamics also change during this
2.4. Erosion Depth Model Based on Dynamic Boundary
process, which affects subsequent stages of erosion damage. The erosion depth obtained
shows Material surface
a dynamic damagetrend
nonlinear induced by erosion is aTo
of accumulation. nonlinear, time-varying
accurately processdamage,
predict erosion
with spatial characteristics. The parameters of particle dynamics also change during this
it is essential to account for the changes in the surface morphology of the flow-passage
process, which affects subsequent stages of erosion damage. The erosion depth obtained
walls
showsover time. nonlinear
a dynamic Therefore,trend
the of
dynamic boundary
accumulation. erosion prediction
To accurately method
predict erosion is utilized
damage,
to calculate the solid–liquid two-phase flow inside the solid–liquid two-phase
it is essential to account for the changes in the surface morphology of the flow-passage flowmeter
in Figure
walls over 3. TheTherefore,
time. dynamicthe boundary
dynamic erosion
boundaryprediction methodmethod
erosion prediction is usedisto dynamically
utilized
calculate
to calculatethe
thewall erosion two-phase
solid–liquid depth based flowon the mesh
inside moving time
the solid–liquid step for
two-phase the flow field.
flowmeter
in Figure
The 3. The dynamic
deformation boundary
of unit surface erosion prediction
(travelling distance) method
is shownis in
used to dynamically
Equation (4):
calculate the wall erosion depth based on the mesh moving time step for the flow field. The
Eis shown
deformation of unit surface (travelling distance) M p  in
tMMEquation (4):
x face  (4)
E × M p × ∆t AMM
face
∆x f ace = (4)
A f ace
T

h = ∑ ∆x
T
h  xface,t (5)
(5)
f ace,t
t 0
t =0

EE represents theDPM
represents the DPM erosion
erosion (m3 /kg);
raterate (m /kg);
3 Mp isMthe
p isparticle
the particle mass∆t(kg);
mass (kg); MM isΔtthe is the mesh
MMmesh
travelling time
travelling step;Aface
timestep; Afaceisisthe mesh
the area
mesh (m2(m
area ); T2);
is T
theis total erosion
the total time; and
erosion h isand
time; the htotal
is the total
erosion depth.
erosion depth.

Figure 3. Flowchart
Figure 3. Flowchartofofprogressive
progressive erosion
erosion prediction
prediction method.
method.
Processes 2024, 12, 1012 5 of 13

3. Mesh Independence and Numerical Calculation Method

3.1. Mesh Independent Verification
To remove the impact of the mesh size on the calculation results, this paper confirmed
the mesh independence of the solid–liquid two-phase flowmeter by measuring the differ-
ential pressure caused by pure oil flowing through the venturi flowmeter, as indicated in
Table 1. An increase in the number of mesh leads to a greater change in the differential
pressure of pure oil flowing through the venturi flowmeter. However, when the mesh
reaches a certain number, the change in differential pressure decreases. Hence, the total
number of mesh cells used in the calculation is 1.55 × 106 .

Table 1. Mesh independent verification.

Mesh Scheme Number of Mesh (106 ) Pressure Difference (Pa)

1 0.77 59,325
2 1.21 59,246
3 1.55 59,181
4 1.64 59,177
5 1.75 59,176

3.2. Boundary Conditions

Based on ANSYS Fluent, the problem of degradation of measuring accuracy caused by
erosion in solid–liquid two-phase flowmeters was calculated using the RNG k-ε to calculate
the fluid phases and the discrete phase model (DPM) to track the particle trajectories. The
velocity inlet was chosen for the inlet of the solid–liquid two-phase flowmeter, while the
pressure outlet was chosen for the outlet. For the inlet of the turbulence term, the turbulence
intensity and hydraulic diameter were selected, and for the outlet, the turbulence intensity
and length scale were chosen. The flow-passage walls were defined as a no-slip wall, and
the particles were set to escape at the inlet and outlet while rebounding from the wall.
Table 2 displays the physical parameters of the wall materials, particles, and crude oil
applied in the numerical simulation of the solid–liquid two-phase flowmeter’s erosion.

Table 2. Physical parameters of wall materials, particles, and crude oils.

Items Physical Parameters Value

Density (kg/m3 ) 7510
Wall materials Vickers hardness (HV) 240
Brinell hardness (HB) 195
Density (kg/m3 ) 2650
Particles
Hardness 9
Density (kg/m3 ) 753.8
Crude Oils
Dynamic Viscosity (kg/(m·s)) 20

3.3. Numerical Calculation Method

The steady-state simulation of flowmeter flow field employs ANSYS FLUENT 2016. In
solid–liquid two-phase flowmeter erosion studies, crude oil is considered an incompressible
fluid, so a pressure-based solver is used. The RNG k-ε model was chosen for the turbulence
model, as shown in Equations (6) and (7).
!
∂ ∂ ∂   ∂k
(ρk) + (ρkui ) = ακ µ e f f + Gk + Gb − ρε − YM (6)
∂t ∂xi ∂x j ∂x j
!
∂ ∂ ∂   ∂ε ε ε2
(ρε) + (ρεui ) = αε µe f f + G1ε ( Gk + G3ε Gb ) − G2ε ρ − R (7)
∂t ∂xi ∂x j ∂x j k k
Processes 2024, 12, 1012 6 of 13

Gk denotes the production of turbulent kinetic energy due to the mean velocity gra-
dient. Gb is the turbulent kinetic energy generated by the buoyancy force. YM denotes
the contribution of fluctuating expansion to the total dissipation rate in compressible tur-
bulence. Ak and αε are the inverse of the effective Prandtl numbers k and ε, respectively.
G1ε = 1.42, G2ε = 1.68.
The control equations were discretized and solved using the SIMPLEC algorithm. The
sub-relaxation factors were set to the default values provided by the system. These values
were as follows: pressure (0.3), density (1), body forces (1), momentum (0.7), turbulent
kinetic energy (0.8), turbulent dissipation rate (0.8), turbulent viscosity (1), and discrete
phase sources (0.5). The pressure, momentum, turbulent kinetic energy, and other factors
were set to the second-order windward format. The criterion for residual convergence
is 1 × 10−8 . The outcomes obtained from the numerical steady state simulation of the
fluid-phase served as preliminary values for the erosion simulation of the multiphase flow.
Then, the discrete phase model (DPM) was applied for calculating the particle erosion.
In the dynamic mesh setup for erosion prediction in solid–liquid two-phase flowme-
ters, the spring smoothing and mesh reconstruction techniques from the smoothing method
were employed to smooth and reconstruct the mesh in the presence of substantial wall
deformation, ensuring the quality of the mesh. After a trial calculation, the initial mesh
movement time step was established as 24 h. In the spring smoothing process, the spring
constant factor was set to 1, while the convergence accuracy and number of iterations
were both left at their default system values, which were 0.001 and 20, respectively. The
element method chose the Tri in Tri Zones. The mesh reconstruction method selected Local
Cell and Local Face, with a minimum mesh size set to 0.4 and mesh size set to 0.9. The
maximum mesh cell skewness was set at 0.8, and the maximum face skewness was set at
0.7. Additionally, the mesh reconstruction frequency was set to 5.

3.4. Experimental Verification

To validate the accuracy of the numerical simulations, this study conducted validation
experiments using the Haimo Technology Group’s test bench, as illustrated in Figure 4.
There is a water reservoir at the back of the test bench, into which prescreened sand with a
volume fraction of 0.09 and diameter of 0.02 mm was introduced into the water reservoir.

(a) (b)

(c)
Figure
Figure 4.
4. Flowmeter
Flowmeter experimental setup. (a)
experimental setup. (a) Water
Water reservoirs;
reservoirs; (b)
(b)flowmeter
flowmetertest
testbench;
bench;(c)
(c)Flowmeter.
Flowme-
ter.
Processes 2024, 12, 1012 7 of 13

Experiments were conducted on the flowmeter at an inlet flow velocity of 4.15 m/s
in accordance with the designed flow rate. The experiments had varying durations of
12 h, 24 h, 36 h, 48 h, and 60 h. After each experiment, the flowmeter was subjected to
ultrasonic cleaning and drying, followed by precise weighing using an electronic scale with
an accuracy of 0.1 g. In this paper, a dynamic boundary erosion prediction method was
employed in numerical simulations. The boundary mesh was reconfigured with erosion
time and erosion rate, resulting in a change in mesh volume. The original mesh body minus
the reconfigured mesh volume represents the lost volume. Finally, the volume multiplied
by the density yields the lost mass. Combining these measurements with the numerical
simulation results, a comparison was made, showing that the numerical results had an error
of less than 3%, as illustrated in Table 3, affirming the reliability of the results presented in
this study.

Table 3. Comparison of calculated and experimental values.

Erosion Time (h) Experimental Value (g) Numerical Result (g) Relative Error (%)
12 2.02 1.98 1.98
24 4.11 4.02 2.19
36 5.97 5.86 1.84
48 8.19 8.05 1.71
60 9.98 10.13 1.47

4. Analysis of Calculation Results

In accordance with the environmental and design requirements for the utilization of a
solid–liquid two-phase flowmeter, the following working conditions have been established.
The particle inflow velocity is 2.1 m/s, 4.2 m/s, 6.3 m/s, and 8.4 m/s. The particle volume
fraction is 0.1, 0.2, 0.3, and 0.5. The particle diameter is 20 µm, 50 µm, 100 µm, and 150 µm.
Experimentally, it was proven that the most serious erosion area of the solid–liquid two-
phase flowmeter was at the tapering tube of the flowmeter, while the erosion area in other
places was negligible. Therefore, the subsequent research will focus on the erosion problem
at the tapering tube.

4.1. Influence of Particle Inflow Velocity on Erosion Evolution Characteristics and

Measurement Accuracy
The erosion depth of flow-passage walls of the solid–liquid two-phase flowmeter for
four particle inflow velocities are displayed in Figure 5 at 2, 4, 6, 8, and 10 years. The erosion
depth curve indicates that the erosion pit depth increases significantly as the erosion time
increases. At an erosion time of 10 years, the maximum erosion depth was 750 µm for a
particle inflow velocity of 8.4 m/s; this depth was probably the maximum erosion depth
for particles at a velocity of 2.1 m/s, which is 2.3 times greater. As the particle flow rate
increases, the peak erosion position changes from 342 mm to 343 mm from the inlet, but
there is no significant difference in the erosion range. Erosion progression leads to a rise in
the quantity of erosion pits and a reduction in the flow velocity within them. Consequently,
particle velocity decreases, reaching its nadir at the maximum depth of erosion. With
decreasing erosion depth and an expanding follow-passage wall of the solid–liquid two-
phase flowmeter, fluid velocity rises. Consequently, particles align with the fluid flow,
resulting in an upsurge in particle velocity.
The flow coefficient is a dimensionless number utilized to characterize the fluid’s
ability to flow through a specific device or channel under defined conditions. It represents
the ratio between the volume of fluid passing through the device or channel per unit of time
and the differential pressure generated during the fluid’s flow, as is shown in Equation (8):
r
ρ/ρ0
Kv = 10Q (8)
∆P
 but there is no significant difference in the erosion range. Erosion progression leads to a
rise in the quantity of erosion pits and a reduction in the flow velocity within them. Con-
sequently, particle velocity decreases, reaching its nadir at the maximum depth of erosion.
With decreasing erosion depth and an expanding follow-passage wall of the solid–liquid
Processes 2024, 12, 1012
two-phase flowmeter, fluid velocity rises. Consequently, particles align with the 8fluid of 13

flow, resulting in an upsurge in particle velocity.

(a) (b) (c) (d)

Figure 5. Erosion depth for different particle inflow velocities over time. (a) up = 2.1 m/s; (b) up = 4.2
m/s; (c) up = 6.3 m/s; (d) up = 8.4 m/s.

The flow coefficient is a dimensionless number utilized to characterize the fluid’s

ability to flow through a specific device or channel under defined conditions. It represents
the ratio between the volume of fluid passing through the device or channel per unit of
time and the differential pressure generated during the fluid’s flow, as is shown in Equa-
(a) (b) (c) (d)
tion (8):
Figure
Figure 5.5.Erosion
Erosiondepth for different
depth particle
for different inflow
particle velocities
inflow
 / velocities
0 over(a)
over time. up = 2.1
time. (a)m/s;
up =(b)2.1
up =m/s;
4.2
m/s; (c) u p = 6.3 m/s; (d) up = 8.4 m/s. K 
(b) up = 4.2 m/s; (c) up = 6.3 m/s; (d) up = 8.4 vm/s. 10 Q (8)
P
The
In flow 6,
InFigure
Figure coefficient
6,when
whenthetheisparticle
a dimensionless
particle inflow
inflowvelocitynumber
velocity utilized
isisless
lessthan to m/s,
than2.1
2.1 characterize
m/s,the flowthe
theflow fluid’s
coefficient
coefficient
ability
increasesto flow through a specific device or channel under defined conditions. It ×
represents −4
increases from 0.9355 initially to 0.9363 over 10 years, reflecting a marginal rise of 8 10
from 0.9355 initially to 0.9363 over 10 years, reflecting a marginal rise of 8 × 10−4
the
withratio
with between
prolonged
prolonged the volume
erosion
erosion time. of fluid passing
time.Nevertheless,
Nevertheless, whenthrough
when the the device
theparticle
particle inflow
inflowor velocity
channel per
fallsunit
velocityfalls withinof
within
the
therange
time and the
range ofof4.2~8.4
4.2~8.4m/s,
differential the
the flow
m/s,pressure coefficient
flowgenerated rises
rises in
coefficientduring conjunction
the
in fluid’s flow,
conjunction with
as is
with both
shown
both erosion time
in Equa-
erosion time
and particle −3−3to
× ×1010
tion particleinflow
and(8): inflowvelocity.
velocity.TheTheincrease
increaseininthe
theflow
flow coefficient
coefficient ranges from
ranges from1.56
1.56 to
− 3
1.97×× 10
1.97 overaaspan
10−3 over spanofof 10
10 years.
years. The
The flow
flow coefficient
coefficient shows
shows aa tendency
tendency toto increase
increasewithwith
both  / 0
botherosion
erosiontime
timeand
andananincrease
increaseininparticle
K v  10inflow
particle Qinflowvelocity.
velocity. (8)
P
In Figure 6, when the particle inflow velocity is less than 2.1 m/s, the flow coefficient
increases from 0.9355 initially to 0.9363 over 10 years, reflecting a marginal rise of 8 × 10−4
with prolonged erosion time. Nevertheless, when the particle inflow velocity falls within
the range of 4.2~8.4 m/s, the flow coefficient rises in conjunction with both erosion time
and particle inflow velocity. The increase in the flow coefficient ranges from 1.56 × 10−3 to
1.97 × 10−3 over a span of 10 years. The flow coefficient shows a tendency to increase with
both erosion time and an increase in particle inflow velocity.

Figure 6. Flow coefficient for different particle inflow velocities over time.

4.2. Influence of Particle Volume Fraction on Erosion Evolution Characteristics and

Measurement Accuracy
In Figure 7, with an increase in particle volume fraction, the erosion depth rises,
reaching a maximum erosion depth of 758 µm after 10 years of erosion, with a particle
volume fraction of 0.5. Compared to a particle volume fraction of 0.1, the maximum erosion
depth is approximately twice as significant. However, the erosion range consistently falls
within the region between approximately 330 mm to 355 mm from the inlet, with the
peak erosion position consistently occurring at approximately 342 mm from the inlet. This
occurs because as the particle volume fraction decreases, the number of particles in the flow
channel also decreases, resulting in reduced erosion damage. Additionally, an increase in
erosion depth elevates the roughness of the follow-passage walls, impeding fluid flow.
 sion depth is approximately twice as significant. However, the erosion range consistently
falls
fallswithin
withinthetheregion
regionbetween
betweenapproximately
approximately330 330mm
mmtoto355
355mm
mmfrom
fromthe
theinlet,
inlet,with
withthe
the
peak
peakerosion
erosionposition
positionconsistently
consistentlyoccurring
occurringatatapproximately
approximately342
342mm
mmfrom
fromthetheinlet.
inlet.This
This
occurs
occursbecause
becauseasasthe
theparticle
particlevolume
volumefraction
fractiondecreases,
decreases,the
thenumber
numberofofparticles
particlesininthe
the
flow
flow channel also decreases, resulting in reduced erosion damage. Additionally,an
channel also decreases, resulting in reduced erosion damage. Additionally, anin-
in-
Processes 2024, 12, 1012 9 of 13
crease
creaseininerosion
erosiondepth
depthelevates
elevatesthe
theroughness
roughnessofofthe
thefollow-passage
follow-passagewalls,
walls,impeding
impedingfluidfluid
flow.
flow.

(a)
(a) (b)
(b) (c)
(c) (d)
(d)
Figure
Figure7.
Figure 7.7.Erosion
Erosiondepth
Erosion depthfor
depth fordifferent
for differentparticle
different particlevolume
volumefractions
volume fractionsover
fractions over
overtime. (a)
time.
time. (a)ααvαv=v=0.1;
(a) (b(b)(b)
=0.1;
0.1; )ααv v=α=v0.2; (c)
(c)
= 0.2;
0.2;
ααv v=α=v0.3;
(c) (d)
= 0.3;
0.3; ααv v=α=v0.5.
(d)(d) = 0.5.
0.5.

InInFigure
Figure8,8,asaserosion
erosiontimetimeincreases,
increases,the
theflow
flow coefficient
flowcoefficient
coefficientexhibits
exhibitsaaconsistent
consistenttrend,
trend,
namely,
namely, an increase in the flow coefficient with the passage of erosion time.
namely, an increase in the flow coefficient with the passage of erosion time. With thepar-
an increase in the flow coefficient with the passage of erosion With
time. Withthe particle
the par-
volume
ticle fraction
ticlevolume
volume increased,
fraction
fraction there was
increased,
increased, a corresponding
there
therewas increase in
wasaacorresponding
corresponding the solid–liquid
increase
increase ininthe two-phase
thesolid–liquid
solid–liquid
flowmeter of 1.04 × 10 −3 , 1.31 ×−310−3 , 1.56 −3× 10−3 , and
two-phaseflow
two-phase flowmetercoefficient,
flow with increments
flowcoefficient, with
withincrements ofof
1.04
1.04××10
10−3, ,1.31
1.31××1010−3, ,1.56
1.56××10
10−3, ,
−3
flowmeter coefficient, increments
1.99 × 10 − 3 over the course of 10 years.
and
and1.99
1.99××10 10−3over
−3
overthe
thecourse
courseofof1010years.
years.

Figure 8.8.Flow
Figure8. coefficient
Flowcoefficient for
coefficientfor different
fordifferent particle
differentparticle volume
particlevolume fractions
volumefractions over
fractionsover time.
overtime.
time.
Figure Flow

4.3. Influence of Particle Diameter on Erosion Evolution Characteristics and

Measurement Accuracy
In Figure 9, with an increase in particle diameter, the depth of the abrasion pit also
increases. At the 10-year mark, the maximum abrasion depth reaches 748 µm when the
particle diameter is 150 µm, which is approximately three times deeper than when the
particle diameter is 20 µm. This is attributed to an escalation in the collision effect of
particles on the wall as the particle diameter increases, resulting in a deeper erosion pit.
With an increase in particle diameter, the abrasion range extends towards the inlet direction.
The distance from the inlet to the flow-passage wall erosion area caused by the four particle
diameters is 333 mm, 331 mm, 328 mm, and 327 mm, respectively. However, the peak
erosion depth remains essentially consistent at approximately 343 mm from the inlet. This
is attributable to the turbulent diffusion effect, where smaller particles, with lower kinetic
energy, exhibit superior adherence to the fluid flow, resulting in a narrower erosion range.
Conversely, larger particles, carrying a higher level of collision energy, generate a more
extensive erosion area. This is primarily attributed to the turbulent diffusion effect: smaller
particles, possessing lower kinetic energy, closely trail the fluid flow, leading to a limited
 The distance from the inlet to the flow-passage wall erosion area caused by the four par-
ticle diameters is 333 mm, 331 mm, 328 mm, and 327 mm, respectively. However, the peak
erosion depth remains essentially consistent at approximately 343 mm from the inlet. This
is a ributable to the turbulent diffusion effect, where smaller particles, with lower kinetic
Processes 2024, 12, 1012 energy, exhibit superior adherence to the fluid flow, resulting in a narrower erosion range.10 of 13
Conversely, larger particles, carrying a higher level of collision energy, generate a more
extensive erosion area. This is primarily a ributed to the turbulent diffusion effect: smaller
particles, possessing lower kinetic energy, closely trail the fluid flow, leading to a limited
extent of erosion damage. Conversely, larger particles carry significant collision energy,
extent of erosion damage. Conversely, larger particles carry significant collision energy,
causing a more extensive area of erosion damage.
causing a more extensive area of erosion damage.

(a) (b) (c) (d)

Figure 9. Erosion depth for different particle diameters over time. (a) dp = 20 µm; (b) dp = 50 µm; (c)
Figure 9. Erosion depth for different particle diameters over time. (a) dp = 20 µm; (b) dp = 50 µm;
dp = 100 µm; (d) dp = 150 µm.
(c) dp = 100 µm; (d) dp = 150 µm.
In Figure 10, when the particle diameter is 20 µm, the increase in the flow coefficient
In Figure 10, when the particle diameter is 20 µm, the increase in the flow coefficient
is the smallest over the course of erosion time, with a gain of 2.644 × 10−2 over 10 years. For
is the smallest over the course of erosion time, with a gain of 2.644 × 10−2 over 10 years.
particle diameters in the range of 50~100 µm, the flow coefficient increases with larger
Forincrements,
particle diameters in the
specifically range
3.156 × 10−2ofand
50~100 × 10the
3.161µm, flow coefficient
−2, respectively. increases
The most with larger
substantial
increments,
increase inspecifically
the flow coefficient 10−2 and
3.156 × occurs when the ×
3.161 10−2 ,diameter
particle respectively.
is 150The
µm,most substantial
reaching a
Processes 2024, 12, x FOR PEER REVIEW 11 of 14
increase
maximum in the flow coefficient
increment occurs
of 3.196 × 10 −2. when the particle diameter is 150 µm, reaching a
maximum increment of 3.196 × 10−2 .

Figure10.
Figure 10.Flow
Flowcoefficient
coefficientfor
fordifferent
differentparticle
particlediameters
diametersover
overtime.
time.

4.4.
4.4.Influence
InfluenceofofParticle
ParticleSphericity
SphericityononErosion
ErosionEvolution
EvolutionCharacteristics
Characteristicsand
and Measurement
Measurement
Accuracy Accuracy
In
In practice, particlesare
practice, particles areseldom
seldomperfectly
perfectly spherical.
spherical. Therefore,
Therefore, investigating
investigating the
the sphe-
sphericity of particles with varying shapes holds significant value for engineering
ricity of particles with varying shapes holds significant value for engineering applications. applica-
tions.
In thisInsubsection,
this subsection, particle
particle sphericity
sphericity is employed
is employed as aasvariable
a variable to elucidate
to elucidate its its impact
impact on
on the erosion damage characteristics and flow coefficient of the solid–liquid
the erosion damage characteristics and flow coefficient of the solid–liquid two-phase flow- two-phase
flowmeter. Particle
meter. Particle sphericity
sphericity is defined
is defined as ratio
as the the ratio of surface
of the the surface
areaarea
of aof a sphere
sphere withwith a
a vol-
volume equal to that of the particle in question to the surface area of the particle itself, as
ume equal to that of the particle in question to the surface area of the particle itself, as
shown in Equation (9):
shown in Equation (9): s
φ= (9)
Ss

In Equation (9), s is the surface area of a sphere of equal volume, and S is the sur- (9)
S
face area of the particle itself. In this paper, three particle sphericities, as proposed by
In Equation
Epstein (9), employed,
[28], are s is the surface area of 0.66,
specifically a sphere
0.76,ofand
equal volume, and S is the surface area
0.86.
of the particle itself. In this paper, three particle sphericities, as proposed by Epstein [28],
are employed, specifically 0.66, 0.76, and 0.86.
In Figure 11, as particle sphericity increases, the depth of erosion pits on the flow-
passage wall of the solid–liquid two-phase flowmeter also increases. The most profound
erosion was noted at 10 years of erosion time, with a particle sphericity of 1, resulting in a
Processes 2024, 12, 1012 11 of 13

In Figure 11, as particle sphericity increases, the depth of erosion pits on the flow-
passage wall of the solid–liquid two-phase flowmeter also increases. The most profound
erosion was noted at 10 years of erosion time, with a particle sphericity of 1, resulting in
a maximum erosion depth of 582 µm. In comparison to the maximum erosion depth at a
particle sphericity of 0.58, this represents an approximately 2.3-fold increase. Nonetheless,
the extent of erosion remains largely consistent, with the peak maximum erosion depth
located at approximately 342 mm from the inlet. With increasing erosion time, the erosion
depth of the solid–liquid two-phase flowmeter’s flow-passage wall consistently rises,
indicating a cumulative effect on erosion depth over time. As erosion depth increases,
particle velocity decreases initially due to the small resistance coefficient of large spherical
particles; however, the gained kinetic energy results in a slight increase in particle velocity.
Processes 2024, 12, x FOR PEER REVIEW 12 of 14
Beyond the erosion peak position, approximately 348 mm from the inlet, as erosion time
increases and the follow-passage wall decreases, fluid velocity rises. This prompts
Processes 2024, 12, x FOR PEER REVIEW 12 of particles
14
to follow the fluid flow, leading to an overall increase in particle velocity.

(a) (a) (b)

(b) (c) (c) (d) (d)
Figure
Figure
Figure 11.Erosion
11.
11. Erosion depth
Erosion depth
depthfor
fordifferent
fordifferentparticle
particle
different sphericity overover
sphericity
particle sphericity time. (a)time.
time.
over φ(a)
= 0.58;=(0.58;
φ(a) bφ) φ
= =0.58;
0.76;
(b (c)
) φ (b) φφ= =(c)
= 0.76; φ=
0.76;
0.86; (d) φ = 1.
(c) φ =(d)
0.86; φ =(d)
0.86; 1. φ = 1.
In Figure 12, as erosion time increases, the flow coefficient of the solid–liquid two-
InInFigure
Figure12,12,as aserosion
erosiontime timeincreases,
increases,the theflow
flowcoefficient
coefficient of ofthe
thesolid–liquid
solid–liquid two-
two-
phase flowmeter exhibits a consistent pa ern across different particle sphericities, demon-
phase
phase flowmeter
flowmeter exhibits a consistent pattern across different particle sphericities, demon-
strating a gradualexhibits
upwardatrend consistent pa ern across
with increasing particledifferent
sphericity.particle sphericities,
Moreover, with a par-demon-
strating
strating aa gradual
gradual
ticle sphericity
upward
upward
of 1,
trend
the flowtrend
withincreasing
with
coefficient
increasingthe particle
experiencesparticle
sphericity.
mostsphericity.
Moreover,
Moreover,
significant increase. As with
with a
par- a par-
particle
ticle sphericity
ticlesphericity
sphericityof of 1,
1, the
rises, the
theflow flow coefficient
flow coefficient experiences
experiences
coefficient exhibits the
the most
a gradual most significant
significant
and consistent increase.
increase.
increase, As
withAs par-
particle
ticle
increasessphericity
sphericity rises,
of 6.9rises,
× 10 the
−4 the
flow
, 1.01 flow coefficient
× 10coefficient
−3 exhibits
, 1.31 × 10 exhibits
−3, and 1.56 a
a gradualgradual
× 10 over
−3
and and consistent
theconsistent increase,
increase,
course of 10 years, with
−4 10−3×, 10 10− 3 , and 1.56 × 10−3 over the course of
with increases
respectively.
increases of 6.9 10×
of ×6.9 −4, 10 10−3×
1.01 ,×1.01 , 1.31 1.31−3, ×
and 1.56 × 10−3 over the course of 10 years,
10 years, respectively.
respectively.

Figure 12. Flow coefficient for different particle sphericity over time.

5. Conclusions
Figure
Figure 12.12.Flow
At Flowcoefficient
coefficient
present, forsimulation
for
numerical differentparticle
different particle sphericity
sphericity
is employed overtime.
over
primarily time.
to investigate the erosion of
fixed boundary conditions or in conjunction with the PIV system, electron microscope
5. scanner,
Conclusions
and so forth to examine the multiphase flow and erosion morphology. However,
these approaches are unablesimulation
At present, numerical to accuratelyis portray
employedthe evolution
primarilyoftoerosion in multiphase
investigate the erosion of
flowmeter flow-passage walls. This paper presents a further study
fixed boundary conditions or in conjunction with the PIV system, electron of the erosion charac-
microscope
scanner, and so forth to examine the multiphase flow and erosion morphology. The
teristics of multiphase flowmeters through the dynamic boundary prediction method. However,
influence of erosion depth change on the average erosion rate and metering accuracy is
these approaches are unable to accurately portray the evolution of erosion in multiphase
analysed, with the results having significant implications for the protection of multiphase
flowmeter flow-passage walls. This paper presents a further study of the erosion charac-
Processes 2024, 12, 1012 12 of 13

5. Conclusions
At present, numerical simulation is employed primarily to investigate the erosion of
fixed boundary conditions or in conjunction with the PIV system, electron microscope scan-
ner, and so forth to examine the multiphase flow and erosion morphology. However, these
approaches are unable to accurately portray the evolution of erosion in multiphase flowme-
ter flow-passage walls. This paper presents a further study of the erosion characteristics of
multiphase flowmeters through the dynamic boundary prediction method. The influence
of erosion depth change on the average erosion rate and metering accuracy is analysed,
with the results having significant implications for the protection of multiphase flowmeter
erosion and improvement of the metering accuracy in practical engineering applications.
(1) Erosion ranges and peak locations on the solid–liquid two-phase flowmeter flow-
passage wall vary with distinct particle dynamic parameters. As the particle flow rate
increases, the erosion range remains largely unchanged. Nevertheless, the erosion
peak shifts to the right, mainly linked to the erosion evolution process. With an
increase in particle diameter, the erosion range extends towards the inlet due to
turbulent diffusion. Particles with lower kinetic energy closely follow the flow. The
erosion range and peak position remained relatively stable despite increases in particle
volume fraction and sphericity.
(2) Erosion depth increases progressively with higher particle dynamic parameters. At a
particle inflow velocity of 8.4 m/s, the maximum erosion depth is 750 µm. Conversely,
at a particle sphericity of 0.58, the minimum erosion depth is 251 µm. Moreover,
a particle volume fraction of 0.5 results in a maximum flow coefficient increase of
1.99 × 10−3 .
This paper explores the influence of dynamic parameters on the multiphase flowmeter
erosion characteristics and measurement accuracy in common working conditions. It is
acknowledged that other conditions and different dynamic parameters may influence
the multiphase flowmeter erosion characteristics and measurement accuracy. Therefore,
further research is required to investigate these factors. In practical engineering, multiphase
flowmeters may encounter two-phase cavitation and cavitation damage in gas–liquid
conditions as well as three-phase abrasion damage in gas–solid–liquid scenarios. Currently,
there is a paucity of research on these issues in multiphase flowmeters. Consequently,
further investigation into the causes and effects of cavitation and erosion damage in
multiphase flowmeters can be conducted at a later stage. Specifically, we will discuss the
potential for future studies to extend our research to include bubbly flow [29] scenarios
with water and solid particles, addressing slipping boundary issues [30] to enhance the
applicability and relevance of our findings in practical settings.

Author Contributions: Conceptualization, W.H. supervised and guided this work. L.Y. developed
the numerical model, ran the simulation, wrote the manuscript, and prepared all the data. Finally, R.L.
and J.Z. reviewed and revised the manuscript before submitting it. X.Y., L.J. and Y.Q. proofread and
typeset the manuscript. All authors have read and agreed to the published version of the manuscript.
Funding: This study was supported by the National Natural Science Foundation of China (Grant
No. 52179086), Central Government Guides Local Science and Technology Development Fund
Projects (Grant No. 23ZYQA0320), Open Research Subject of Key Laboratory of Fluid Machinery and
Engineering (Xihua University), Sichuan Province (Grant No. LTJX-2023003), and Natural Science
Foundation of Gansu Province (Grant No. 23JRRA788). Thanks to the staff of Haimo Technology
Group for the help.
Data Availability Statement: All data in this manuscript are available from the corresponding author
by e-mail.
Conflicts of Interest: The authors declare no conflict of interest.
Processes 2024, 12, 1012 13 of 13

References
1. Li, F. Innovative applications of multiphase mixing pumps. Intern. Combust. Engine Parts 2019, 05.100, 203–204.
2. Feng, D.; Yin, S.; Wang, P. Research Progress of Downhole Flow Measurement and Controlling Technique. J. Oil Gas Technol. 2007,
29, 148–150+170.
3. Liu, M.; Tan, L.; Cao, S. Dynamic mode decomposition of gas-liquid flow in a rotodynamic multiphase pump. Renew. Energy 2019,
139, 1159–1175. [CrossRef]
4. Rabha, S.; Schubert, M.; Grugel, F.; Banowski, M.; Hampel, U. Visualization and quantitative analysis of dispersive mixing by a
helical static mixer in upward co-current gas–liquid flow. Chem. Eng. J. 2015, 262, 527–540. [CrossRef]
5. Suh, J.W.; Choi, Y.S.; Kim, J.H.; Lee, K.Y.; Joo, W.G. Multiphase flow analysis for air-water bubbly flow in a multiphase pump.
In Proceedings of the ASME 2017 Fluids Engineering Division Summer Meeting, Waikoloa, HI, USA, 30 July–3 August 2017;
American Society of Mechanical Engineers: New York, NY, USA, 2017; Volume 58042, p. V01AT03A015.
6. Yu, F.; Ren, A.; Liu, K. Research statue of helical axial-flow pump for multiphase transfer. Oil Field Equip. 2004, 4–7.
7. Li, M.; Li, X.; Zhang, X. Optimization of Multiphase Flowmeter on Offshore Oil Production Platform. Mod. Chem. Res. 2020, 60,
131–132.
8. Gu, S.; Wang, W. Discussions on The Measurement of Gas-Solid Flow by Means of Venturi meter. J. Southeast Univ. (Nat. Sci. Ed.)
1984, 64–71.
9. Chen, L.; Gong, D. Measurement of moisture using differential pressure flow meters. Oil-Gas Field Surf. Eng. 2013, 32, 31–32.
10. Feng, S.; Guo, K. Oil and Gas Gathering and Mine Processing; China University of Petroleum Press: Beijing, China, 2006.
11. Ma, Y.; Zheng, J.; Tang, X. Research on the application of multi-phase flowmeter in Bohai Sea thick oil field. Pet. New Energy 2012,
23, 36–41.
12. He, D.; Bai, B. A new correlation for wet gas flow rate measurement with Venturi meter based on two-phase mass flow coefficient.
Measurement 2014, 58, 61–67. [CrossRef]
13. Li, W.; Li, Z.; Deng, W.; Ji, L.; Qiu, Y.; Chen, H. Particle image velocimetry flowmeter for natural gas applications. Flow Meas.
Instrum. 2021, 82, 102072.
14. Li, W.; Li, Z.; Han, W.; Li, Y.; Yan, S.; Zhao, Q.; Chen, F. Measured viscosity characteristics of Fe3 O4 ferrofluid in magnetic and
thermal fields. Phys. Fluids 2023, 35, 012002. [CrossRef]
15. Wang, S.; Liu, H.; Zhang, R.; Liu, M. Numerical simulations of sand erosion in pipelines and evaluations of solid particle erosion
equations. Ocean. Eng. 2013, 32, 49–59.
16. Singh, N.K.; Kumar, A.; Ang, A.S.; Mahajan, D.K.; Singh, H. Characterization and slurry erosion mechanisms of Nickel-Based
cermet coatings on monel K-500. J. Therm. Spray Technol. 2021, 30, 2138–2154. [CrossRef]
17. Zhao, W.; Guo, Q.; Song, Q. Effects of Blade Entrance Geometrical Parameter on Centrifugal Pump Wear Rule. J. Gansu Sci. 2018,
30, 126–131.
18. Desale, G.R.; Gandhi, B.K.; Jain, S.C. Particle size effects on the slurry erosion of aluminium alloy. Wear 2009, 266, 1066–1071.
[CrossRef]
19. Liu, Z.; Wan, S.; Nguyen, V.B.; Zhang, Y. A numerical study on the effect of particle shape on the erosion of ductile materials.
Wear 2014, 313, 135–142. [CrossRef]
20. Nguyen, Q.B.; Nguyen, V.B.; Lim, C.; Trinh, Q.T.; Sankaranarayanan, S.; Zhang, Y.; Gupta, M. Effect of impact angle and testing
time on erosion of stainless steel at higher velocities. Wear 2014, 321, 87–93. [CrossRef]
21. Liu, H.; Wei, N. Study on erosion damage of solid particles to oil and gas well tubing. China Min. Mag. 2018, 27, 157–162.
22. Li, Z.; Du, M.C.; Dong, X. Design and tests for single angular particle erosion mechanism test system. J. Vib. Shock. 2018, 37, 195–203.
23. Gajan, P.; Decaudin, Q.; Couput, J.P. Analysis of high pressure tests on wet gas flow metering with a Venturi meter. Flow Meas.
Instrum. 2015, 44, 126–131. [CrossRef]
24. Dehkordi, P.B.; Colombo LP, M.; Guilizzoni, M.; Sotgia, G. CFD simulation with experimental validation of oil-water core-annular
flows through Venturi and Nozzle flow meters. J. Pet. Sci. Eng. 2017, 149, 540–552. [CrossRef]
25. Liu, G.; Li, J.; OuYang, Z. Numerical analysis of flow performance on small-diameter cryogenic Venturi flowmeter Cryogenics.
Low Temp. Eng. 2021, 240, 41–46.
26. Oka, Y.I.; Okamura, K.; Yoshida, T. Practical estimation of erosion damage caused by solid particle impact: Part 1: Effects of
impact parameters on a predictive equation. Wear 2015, 259, 95–101. [CrossRef]
27. Oka, Y.I.; Yoshida, T. Practical estimation of erosion damage caused by solid particle impact: Part 2: Mechanical properties of
materials directly associated with erosion damage. Wear 2005, 259, 102–109. [CrossRef]
28. Epstein, N. Handbook of Fluidization and Fluid-Particle Systems; CRC Press: Boca Raton, FL, USA, 2003.
29. Li, Z.; Li, J.; Yan, G.; Galindo-Torres, S.; Scheuermann, A.; Li, L. Mesoscopic model framework for liquid slip in a confined
parallel-plate flow channel. Phys. Rev. Fluids 2021, 6, 034203. [CrossRef]
30. Ma, X.; Gu, Z.; Ni, D.; Li, C.; Zhang, W.; Zhang, F.; Tian, M. Experimental Study on Gas–Liquid Two-Phase Flow Upstream and
Downstream of U-Bends. Processes 2024, 12, 277. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.