Engineering Applications of Computational Fluid

Mechanics

ISSN: 1994-2060 (Print) 1997-003X (Online) Journal homepage: https://www.tandfonline.com/loi/tcfm20

Influence of inlet pressure on cavitation

characteristics in regulating valve

Xiumei Liu, Zihong Wu, Beibei Li, Jiyun Zhao, Jie He, Wei Li, Chi Zhang &
Fangwei Xie

To cite this article: Xiumei Liu, Zihong Wu, Beibei Li, Jiyun Zhao, Jie He, Wei Li, Chi Zhang
& Fangwei Xie (2020) Influence of inlet pressure on cavitation characteristics in regulating
valve, Engineering Applications of Computational Fluid Mechanics, 14:1, 299-310, DOI:
10.1080/19942060.2020.1711811

To link to this article: https://doi.org/10.1080/19942060.2020.1711811

© 2020 The Author(s). Published by Informa

UK Limited, trading as Taylor & Francis
Group

Published online: 12 Jan 2020.

Submit your article to this journal

Article views: 101

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=tcfm20
ENGINEERING APPLICATIONS OF COMPUTATIONAL FLUID MECHANICS
2020, VOL. 14, NO. 1, 299–310
https://doi.org/10.1080/19942060.2020.1711811

REVIEW ARTICLE

Influence of inlet pressure on cavitation characteristics in regulating valve

Xiumei Liu, Zihong Wu, Beibei Li, Jiyun Zhao, Jie He, Wei Li, Chi Zhang and Fangwei Xie
School of Mechatronic Engineering, China University of Mining and Technology, Xuzhou, People’s Republic of China

ABSTRACT ARTICLE HISTORY

The cavitation development process in regulating valve under different inlet pressures is investi- Received 16 June 2019
gated experimentally and numerically in our paper. The influence of inlet pressure Pin on cavitation Accepted 28 December 2019
location, shape, area and intensity is studied in axial and radial directions, and the axial and radial KEYWORDS
pressure distributions are also analyzed. The cavitation is generally annular in shape radially and Regulating valve; cavitation
irregular polygon in shape axially. Cavitation bubbles occur at the throat entrance when the inlet flow; bubble ring; inlet
pressure increases. The bubbles then accumulate, and a bubble ring is formed. The cavitation devel- pressure; cavitation area;
opment process under different inlet pressures is separated into three stages, no-cavitation, initial cavitation intensity
cavitation and steady-cavitation stages. During the no-cavitation stage, no bubbles exist. During the
initial cavitation stage, bubbles occur and the area and intensity of cavitation increase in the radial
and axial directions with increasing inlet pressure. During the steady-cavitation stage, the calculation
area and intensity of bubble rings reach their maximum values and then remain stable. The cavitation
bubbles extend downstream, and the area of cavitation ring and cavitation intensity increase with
increasing inlet pressure. A slit exists between the upper and lower parts of bubbles in the cavitation
process, both parts touch each other forming a whole.

1. Introduction
Ma (2017) investigated internal flow characteristics, and
A regulating valve is a key part of direct coal liquefaction, discussed the flow rate under varying opening degrees,
which often suffers from severe shocks, cavitation erosion which is affected by cavitation. Forder, Thew, and Har-
and damage (Han, Liu, Wu, Zhao, & Tan, 2017; Liu, Jiang, rison (1998) presented a numerical erosion model that
& Zheng, 2012; Zheng, Ou, Ye, et al., 2016). Surveys show could predict erosion rates, and they indicated that veloc-
that the lifetime of a regulating valve is less than half a ity was a major reason for erosion. Jin, Gao, Zhang, and
year, which seriously affects safety performance and the Qian (2017) emphasized that inlet velocity has an impor-
long-term stable operation of coal chemical systems (Jin tant effect on pressure drop and determined the pressure
et al., 2015). The flow characteristics of fluid in the reg- distribution with variable inlet velocities. Liu, Zhao, and
ulating valve have a significant effect on its performance. Qian (2017) analyzed the flow characteristics, such as
Therefore, understanding the complex flow characteris- water volume fraction, flow rate and turbulent intensity,
tics inside the valve thoroughly is important, and the and then obtained the effect of these factors on the flow
details of flow characteristics are helpful for optimizing features of fluid.
the valve structure. Many researchers have discussed the flow perfor-
Many studies have focused on the main factors that mance under different inlet boundaries and its different
influence flow characteristics, such as opening degree, developing process (Long et al., 2017; Yang, Wang, & Lu,
medium, velocity, geometry configuration, and pressure 2019). Chern, Wang, and Ma (2007) visualized the per-
difference (Akbarian et al., 2018; Amirante, Distaso, formance and pattern of flow inside a valve with differ-
& Tamburrano, 2014; Ramezanizadeh, Alhuyi Nazari, ent opening degrees and inlet velocities experimentally.
Ahmadi, & Chau, 2019; Zheng, Ou, Ye, et al., 2016). Mou, The cavitation index was defined to predict the incep-
He, Zhao, and Chau (2017); Mosavi, Shamshirband, Sal- tion of cavitation based on experimental data. Abdulaziz
wana, Chau, and Tah (2019); and Ghalandari, Mirzadeh (2014) captured the dynamic behavior of cavitation bub-
Koohshahi, Mohamadian, Shamshirband, and Chau bles under different pressure ratios and obtained the
(2019) studied internal flow characteristics based on change in cavitation morphology. Yuan, Song, and Liu
computational fluid dynamics. Cui, Lin, Zhu, Wang, and (2019) observed the morphologies of cavitation bubbles

CONTACT Zihong Wu wuzihongcumt@163.com; Fangwei Xie xiefangwei@163.com

© 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.
300 X. LIU ET AL.

and identified the growth process of bubbles from cav-

itation nuclei to bubble cluster with increasing pressure
drop.
The literature has focused on the characteristics of
flow with different operating conditions, but few reports
exist on the cavitation characteristics in regulating valves
used in direct coal liquefaction. Only a few papers
(Zheng, Ou, & Jin, 2017; Zheng, Ou, Ye, et al., 2016)
reported the cavitation erosion of high-pressure letdown
valve. They investigated cavitation-flow-induced valve
spool failure, and a cavitation flow test rig was developed
to observe cavitation on the valve model. However, the
valve model in their papers was not strictly similar to
the actual geometry but was only a combination of ven-
turi and turbulent flow. The internal flow structure of the
regulating valve is more complicated than that in their
papers. Understanding the cavitation flow in the regu-
lating valve under different inlet pressures is important
owing to the noticeable differences. A 3D computation
model is built, and a visualization experimental method Figure 1. (a) Schematic of experimental setup; (2) Geometrical
is designed to capture cavitation flow characteristics with structure of regulating valve.
different inlet pressures in the regulating valve. A trans-
parent model valve with a flow channel similar to the 3. Numerical simulation
actual geometry in every aspect is machined in our test;
thus, the cavitation development process is close to the The commercial software Ansys Fluent 15.0 is employed
flow field in a real scale. In addition, the cavitation flow for numerical calculation of the cavitation flow in the
under different inlet pressures is presented in our paper, regulating valve. The fluid inside the regulating valve is
and the cavitation area and length are determined by a complex multiphase flow (Zheng, Ou, Ye, et al., 2016;
image processing. Not only the cavitation bubble loca- Zheng, Ou, Yi, et al., 2016), and a transformation from
tion and cavitation shape (the outer extent of cavitation liquid to vapor exists when bubbles occur. Based on the
zones) but also the cavitation bubble ring and cavitation Navier–Stokes equation, the gas–liquid phase is consid-
development process under different inlet pressures Pin ered continuous, and the mixture model is often selected
in axial and radial directions are discussed in our paper. to describe multiphase flow (Qiu et al., 2018b; Yin, Yu,
Jia, & Yu, 2016).

2. Experimental investigation system 3.1. Governing equations

The instruments and experimental setup are presented The continuity equation for the mixture is:
in Figure 1(a). The pressure gauge is used to measure ∂
the inlet and outlet pressures, and the turbine flowme- (ρm ) + ∇ · (ρm v) = 0 (1)
∂t
ter upstream is used to measure liquid flow. The geo-
metrical structure of the regulating valve is designed in The momentum equation for the mixture can be
accordance with the actual one used in direct coal lique- expressed as:
faction. The size of the transparent valve is at one-tenth ∂
scale of the original geometry, which is made of poly- (ρm v) + ∇ · (ρm vv) = −∇p + ∇[μm (∇v + ∇vT )]
∂t
methyl methacrylate and transparent for good visualiza- (2)
tion observation. The experimental valve has a length
of 77 mm, a width of 25 mm and a minimum diameter Here, μm is the viscosity of the mixture; ρm is the mix-
of 3 mm, as shown in Figure 1(b). An optical fiber light ture density defined as ρm = ρv αv + ρl (1 − αv ); ρv ,ρl
(LA-100USW) is used as the illumination light, and a are the vapor and liquid densities, respectively; v is the
high-speed camera is used to observe the cavitation flow mass-averaged velocity; αv , 1 − αv are the vapor and liq-
(Phantom VEO-710L). The optical fiber source and high- uid volume fractions, respectively; and ∇p is the gradient
speed camera are mounted on the same side. of pressure p.
 ENGINEERING APPLICATIONS OF COMPUTATIONAL FLUID MECHANICS 301

The net mass transfer from liquid to vapor is often

described by the Schnerr–Sauer cavitation model, which
can be expressed as:
 
ρl ρv 3 2 Pv − P 1/2
Re = αv (1 − αv ) ; P ≤ Pv (3)
ρm RB 3 ρl
 
ρl ρv 3 2 P − Pv 1/2
Rc = αv (1 − αv ) ; P ≥ Pv (4)
ρm RB 3 ρl

In the formula, RB is the bubble radius, RB =

 1
αv 3 1 3
1−αv 4π n ; P and Pv is the cell center pressure and sat-
uration vapor pressure; and n is the number of bubbles
per volume of liquid, and the model takes n = 1013 .
In our paper, the Reynolds number ranges from 1052
to 1590, which is much larger than the critical Reynolds
number of cone valve; the standard k- turbulence model
is suggested to describe turbulence effects (Andriotis,
Gavaises, & Arcoumanis, 2008; Margot, Hoyas, Gil, & Figure 2. (a) Axisymmetric model of regulating valve; (b) Mesh
Patouna, 2014). The bubbles’ movement greatly affects modeling.
the liquid-phase turbulent kinetic energy, and dissipation
effect is also considered (Laı´n, Bröder, Sommerfeld, &
Göz, 2002). The mixture, standard k- turbulence and
Schnerr–Sauer models are used in our calculation. isothermal and incompressible. The density is 889 kg/m3 ,
and the viscosity is 4.6 × 10−5 (m2 /s).

3.2. Geometry and mesh modeling of regulating

valve 3.3. Grid independence analysis
Figure 2(a) shows the axisymmetric model of the regu- The grid independence analysis is depicted in Figure 3
lating valve, which consists of a spool and a valve seat. when Pin = 2.5 MPa and Pout = 1 MPa. The data show
Three different sections of the flow channel exist in this that the mass flow rate increases rapidly with increas-
regulating valve, which are called contraction, throat and ing grid numbers. When the grid number is greater
diffuser sections. We can change the opening degree of than 200,000 and the minimum size is 0.08*0.08 mm,
the regulating valve by changing the relative position of the curve of mass flow rate remains stable. The tendency
the spool and valve seat. The flow direction is consistent of mass flow rate demonstrates that the results are grid-
with the positive X direction. The computed geometry is independent. Thus, the amount of grid 200,000 is used
discretized by using the hexahedron mesh, as shown in in our simulation to save computation time, whilst the
Figure 2(b). The throttling zone has the finest mesh den- precision is guaranteed.
sity due to the great pressure or velocity gradients caused
by the contracted flow area. Encryption in the throttling
zone is important.
The relative opening degree of the regulating valve is
defined as:
l
K= (5)
lmax
where l is the displacement of the regulating valve at a cer-
tain opening, and lmax is the total stroke of the regulating
valve.
The inlet boundary condition is defined using inlet
pressure (Pin ), and the outlet boundary condition is
defined using outlet pressure (Pout ). No. 46 antiwear
hydraulic oil is used in our system, which is assumed to be Figure 3. Grid independence.
302 X. LIU ET AL.

4. Result
When cavitation bubbles occur, almost all lights are
reflected at the interface between the cavitation bub-
bles and oil. The corresponding zones of the bubbles are
bright, whereas those of the oil are dark. Thus, the cav-
itation and non-cavitation bubbles can be easily distin-
guished. The digital image-processing techniques based
on gray-level detection on the images are widely used in
two-phase mixture identification (de Graaf, Brandner, &
Pearce, 2017; Soyama & Hoshino, 2016), and the prin-
ciple of calculating cavitation bubble length is shown in
Figure 4. Firstly, the image is tailored to the proper area
to reduce the number of pixels needed to be calculated.
The average gray value G is determined by calculating the
gray value of the same abscissa, and an average value is
obtained. Then, the gray value λ is defined as:

G
λ= (6)
256

Figure 4(b) shows the variation in the gray value λ.

A criterion should be introduced to calculate the cavita-
tion because the cavitation boundary is distinct but not
neat. As discussed by (Wang et al., 2018), the average of
the gray value λs = λmax +λ
2
min
is employed as the thresh-
old, which is used to define the initial and the collapse
end point of the cavitation. If the distance λ ≥ λs , then
the distance from the collapse to start boundary is the
cavitation length.

Figure 5. Growing trend of cavitation length under different Pin

(Pout = 1 MPa).

Figure 5 shows the growing trend of cavitation length

under increasing Pin at a relative opening degree of
50%. The images in the left column are the experi-
mental results (A1 –A6 ) captured by a high-speed cam-
era, and those in the right column are the correspond-
ing numerical results (B1 –B6 ). When the backpressure
is set as 1 MPa, cavitation bubbles occur at the throat
entrance and develop downstream with increasing Pin .
When the oil flows to the contraction section of the
flow passage, the flow direction changes because of the
decrease in geometric area, which induces increasing
velocity and decreasing liquid pressure (Yin et al., 2016).
If the liquid pressure is below the saturated vapor pres-
sure (Pv ), then cavitation incepts can be found (Qiu et al.,
2016b). Based on the digital image-processing methods
mentioned, the white dot in image A1 shows the start-
ing position of cavitation, and the moving distance of
Figure 4. The method of cavitation length calculation. white dots in images A2 –A6 shows the changing trend of
 ENGINEERING APPLICATIONS OF COMPUTATIONAL FLUID MECHANICS 303

cavitation length. The numerical results imply that cavi-

tation length increases as Pin increases from 2.1 to 4 MPa.
Figure 5 also demonstrates that the value of vapor volume
fraction becomes larger with increasing Pin . The cavita-
tion bubbles become stronger when the input pressure
increases. The reasons are as follows: the pressure dif-
ference increases as Pin rises and a monotonic increase
in the mass flow rate can be found (Qiu et al., 2016b).
The larger the mass flow rate is, the more energy is
converted into kinetic energy, thereby inducing more
cavitation bubbles. With increasing Pin , the high-speed
zones near the orifice and at the contraction section exit
expands the scope, causing larger low-pressure regions.
Then, a larger amount of dissolved gas nuclei in the low-
pressure regions are formed to the cavitation bubbles.
Figure 7. The flow rate obtained from simulation and the exper-
Furthermore, increasing input pressure generates more imental results.
flow vortexes and much stronger turbulence in the chan-
nel (Wei et al., 2017). The highly strengthened vortexes
and more intense mixing rate in the valve could cause
stronger cavitation (Dai et al., 2019). Similar phenomena obtained from the experimental and numerical results.
can be found elsewhere in the literature (Dai et al., 2019; When Pin increases from 2.1 to 4 MPa, the flow rate
Qiu et al., 2016b; Wei et al., 2017). increases from 26 to 43.3 L/min. Cavitation bubbles
Figure 6 shows the cavitation length obtained from occupy part of the flow passage, thereby reducing the
Figure 5, which indicates that the cavitation length actual flow area greatly and increasing the pressure loss.
increases with increasing Pin . Good agreement between The experimental and simulation results have good con-
the predicted and measured lengths is obtained. The sistency, and the maximum error between them is 2%.
maximum relative error between the simulation and Comparison of the cavitation lengths and flow rates
experimental data is approximately 3%. This error is obtained by simulation and experimental methods shows
probably due to the non-condensable gases not consid- that the model in our study proves its correctness for
ered in our simulation model. Several factors, such as use to investigate the flow characteristic of cavitation in
liquid compressibility, temperature rise and machining regulating valves.
tolerance, are also among the main reasons for the dif-
ferences between the computed results and experimental
data. 5. Discussion
Figure 7 shows the variation in flow rate with increas-
5.1. Cavitation development process in radial
ing Pin . When Pin increases, the flow rate increases, as
direction
Figure 8(a) illustrates the cavitation development stage
at the throat of the regulating valve with variable inlet
pressures. Bubbles appear near the throat, grow along the
inner surface of the throat and form bubble rings. Five
planes (named A, B, C, D, and E), which are spaced with
equal distances from one another, are shown in Figure 8
to discuss the distribution of vapor volume fraction radi-
ally. The average intervals are 1.5 mm. The bubbles can
be found on the planes, and bubble rings occur. The
bubble rings are attached to the inner and outer bound-
aries of the plane and extend towards its center. With
increasing Pin , bubbles could also be found on planes B,
C, D, and E. The vapor volume fraction on each plane
becomes stronger with rising Pin . In the radial direction,
Figure 6. Cavitation length obtained by numerical and experi- the vapor–liquid interface in the planes is close to each
mental methods. other and finally intersects.
304 X. LIU ET AL.

Figure 8. Numerical results of cavitation development process under different inlet pressures (a) inside the regulating valve, (b) on plane
C, and (c) on five different planes.

To interpret the cavitation development process, a new central position of five planes. The cavitation develop-
coefficient Am being used is expressed as. ment process can be separated into three periods (Qiu,
Wang, Lei, Wu, & Guo, 2018a; Qiu et al., 2018b). Period

N
I is the no-cavitation stage (Pin ≤ 2.5 MPa) in which no
Am = Ai (7)
bubbles occur (Am = 0). Period II is the initial cavita-
i=1
tion stage (2.5 MPa < Pin ≤ 3.5 MPa). During this stage,
Ai is defined as the region of vapor volume fraction that the curve of Am rises with increasing Pin. The bubble
is over 1% in the radial direction (Li, Liu, Wei, Liang, & rings seem to increase with the accumulation of cavita-
Tang, 2018; Qiu et al., 2016a). Figure 8(b) shows the rela- tion bubbles. Oil and vapor mixture can be found inside
tionship between Am and Pin on plane C, which is the the bubble rings. The bubbles grow in the inner and outer
 ENGINEERING APPLICATIONS OF COMPUTATIONAL FLUID MECHANICS 305

Figure 9. Numerical results: (a) model schematic of cavitation flow in the valve (not to scale), (b) cavitation intensity on plane A, (c)
cavitation intensity on plane C, and (d) cavitation intensity on plane E.

boundaries of plane C, extend towards one another and occur and attach on the inner surface of the throat and the
finally intersect in the middle as Pin increases. Coefficient outer surface of the spool head, as shown in Figure 9(a).
Am and the vapor volume fraction on plane C increase as Then, the flow velocity decreases, and the pressure grad-
Pin rises. The cavitation becomes stronger with increas- ually recovers with a wide flow channel. The pressure
ing input pressure (Qiu et al., 2016a; Qiu et al., 2016b). gradient is contrary to the flow direction (Jin et al., 2015).
Period III is called the steady cavitation (Pin > 3.5 MPa), With this adverse pressure gradient, backflow zones can
in which coefficient Am and the vapor volume fraction be found around the top of the head of the valve spool as
on plane C reach their maximum value and remain sta- a result of fluid viscosity. As affected by the high-speed
ble even if Pin continues to increase. Figure 8(c) shows backflow, the area of the upper bubbles is larger than that
coefficient Am on planes A, B, C, D, and E with increas- on the lower side (Zheng, Ou, Yi, et al., 2016).
ing Pin . The bubbles appear in sequential order on planes The density of bubbles in the cavitation region (cav-
A, B, C, D, and E as Pin increases. The farther away from itation intensity) can also be represented by using the
the throttling zone, larger Pin is needed for the cavitation vapor volume fraction on planes A, C, and E, as pre-
inception. The cavitation inception occurs on plane A as sented in Figure 9(b–d). If bubbles occur, then the vapor
Pin reaches 2.1 MPa. On plane D or E, the critical pres- volume fraction on plane A decreases firstly, and then
sure of the cavitation inception rises to Pin = 3.0 MPa. remains stable and finally increases along the Y-axis. This
Am on planes rises to its maximum value and stabilizes condition illustrates that the cavitation intensity on the
with rising Pin . The maximum value of Am is 1.83 mm2 surface of the throat and spool head is the strongest,
when Pin = 2.1 MPa. When Pin = 4.0 MPa, the maxi- and the minimum value is located near the slit between
mum value of Am on plane E is greater than 17. The plane the valve spool and seat. The reason is that cavitation
far from the throttling zone is covered by substantial is firstly generated on the surfaces and then developed
bubbles, and the effective flow area reaches its minimum. into the gaps. With Pin ≥ 2.5 MPa, the distribution of
As shown in Figure 5, the bubbles occur near the cavitation intensity and area on plane A occurs in the
throttling orifice and attach on the walls of contraction same manner, which shows that the cavitation process
and spool. The fluid accelerates through the contraction comes into the steady-cavitation stage. The distribution
section, and the pressure gradient is consistent with the of volume fraction on plane C is a typical case rep-
flow. When the local pressure is lower than Pv , bubbles resenting the entire cavitation development process, as
306 X. LIU ET AL.

shown in Figure 9(c). When Pin ≤ 2.5 MPa, it is in the

no-cavitation stage because the pressure is higher than
Pv . When Pin = 3.0 MPa, the change in cavitation inten-
sity on plane C is complex, namely, firstly rises, then
falls and finally rises again. The first rising point locates
at the reflux region and the end of the lower bub-
bles. With increasing Y-coordinates, cavitation intensity
decreases, which corresponds to the position of the chan-
nel center. Subsequently, the position extends to the inte-
rior of the upper bubbles. The vapor volume fraction
rises again and reaches the maximum value on the sur-
face of the contraction section. When Pin > 3.0 MPa, the
vapor volume fraction reaches the maximum near the
walls and the minimum values locate near the gap. The
change trend of vapor volume fraction is almost the same,
which shows that the cavitation process has entered the
steady cavitation stage, as discussed above. Figure 9(d)
shows the developing process of vapor on plane E. When
Pin = 3.5 MPa, the fluctuation in vapor volume fraction
is influenced by the upper bubbles. The curves of vapor
volume fraction at Pin = 4.0 MPa on plane E have a sim-
ilar tendency with that at Pin = 3.5 MPa on plane C. This
condition demonstrates that cavitation bubbles are in the
initial cavitation stage with a developing process similar
to each other.
Figure 10(a,b) show the distribution of pressure in the
throat along the surface of the valve seat and the spool
head under different inlet pressures. When the oil flows
to the contraction section with decreasing flow area,
the pressure decreases and the velocity increases. With
increasing Pin , the pressure drop gradient in the contrac-
tion section increases. In Figure 10(a), Pin is 2.1 MPa,
and the minimum pressure along the valve seat sur-
face on the throat surface is larger than Pv . When Pin
increases to 2.2 MPa, the pressure is lower than Pv , and
cavitation occurs in the throat. Considerable energy is
Figure 10. Numerical results of pressure distribution in the
converted into kinetic energy with increasing Pin , and throat: (a) along the valve seat surface, and (b) on the surface of
high-speed and low-pressure zones enlarge. Thus, the the spool head.
larger the input pressure is, the larger the low-pressure
area becomes. Figure 10(b) shows the pressure distribu-
tion on the outer surface of the head of the valve spool,
5.2. Cavitation development process in axial
which is similar to the distribution of pressure along the
direction
valve seat surface. The pressure drop mainly happens at
the orifice with the smallest area. Whilst the oil moves The area of vapor volume fraction greater than 1% on
downstream, the flow area increases again, the velocity the rotating surface is defined as Ac in the axial direc-
decreases and the pressure rebounds slightly. The pres- tion. Figure 11(a) presents the cavitation isograms with
sure rebound is located near the throat exit with high vapor volume fraction equal to 1%. When Pin < 2.2 MPa,
Pin . If Pin rises up to 4.0 MPa, the spool head is cov- no bubbles occur. At Pin = 2.2 MPa, cavitation can be
ered by low pressure. Consequently, no pressure rebound found, and a slit exists between the upper and lower parts
phenomenon occurs on the spool surface. The pressure of the cavitation bubble. The upper and lower parts of
distribution fluctuates on the surface of the spool head bubbles touch each other and form a concave geometry
in the throat due to the geometric characteristics of the later when Pin = 2.5 MPa. With the continuous develop-
spool head. ment of the cavitation process, the value of Ac becomes
 ENGINEERING APPLICATIONS OF COMPUTATIONAL FLUID MECHANICS 307

Figure 11. Numerical results: (a) cavitation isograms with 1% vapor volume fraction and (b) influence of Pin on Ac .

larger. The increase rate of cavitation area becomes larger, the rising tendency of vapor volume fraction on the spool
and the cavitation bubble propagates downstream, when head under different inlet pressures is smooth, which
Pin further increases. When Pin increases to 4.0 MPa, the illustrates that the growing degree of the cavitation inten-
cavitation bubbles nearly fill the throat of the regulating sity is slower than that on the surface of the valve seat.
valve and form a smooth outline shape in the flow passage Nonetheless, a sudden drop in cavitation intensity hap-
(Long et al., 2017). pens on the inner surface of the valve seat and the outer
Figure 12 shows the distribution of cavitation intensity surface of the spool head. The cavitation intensity starts
or vapor volume fraction on the throat surface. Whilst to increase near the throttling zone and rises to its maxi-
Pin increases, the maximum of the vapor volume fraction mum values at the position of the minimum effective flow
becomes higher, and the length of the vapor–liquid mix- cross-section. The fluctuation in drop line, as shown in
ing section is also extended. Substantial pressure energy circles in Figure 12(b), is due to the unsmooth transition
is converted into kinetic energy, and the maximum value of contour lines on the spool head.
of flow velocity increases with increasing Pin . Therefore, Figure 13 shows the maximum velocity under vari-
the high-velocity zone has the tendency of radial expan- able inlet pressures obtained numerically when the Schn-
sion and the velocity gradient between the valve seat and err–Sauer cavitation model is employed or not. The max-
the center of the flow passage increases. This larger veloc- imum velocity rises with increasing Pin because substan-
ity gradient enhances the shear strength, the maximum of tial pressure energy is converted into kinetic energy. A
the vapor volume fraction becomes larger (Zhang et al., velocity difference exists between the single and mixing
2014), and the cavitation strengthens with increasing fluid. The maximum velocity is smaller when the cavita-
input pressure. When Pin = 2.1 MPa, no cavitation can tion model is employed. As the fluid flows through the
be found, and the vapor volume fraction equals 0. As Pin contraction section of the valve, the bubbles occur once
approaches 4.0 MPa, the peak value of the vapor volume the pressure reduces to the saturated pressure. The re-
fraction at 33.2 mm is close to 0.95, which demonstrates entrant flow near the end of the cavitation bubbles greatly
that the inner surface of the throat is almost covered by increases the turbulence intensity, consuming a substan-
pure vapor. Figure 12(b) shows the cavitation intensity tial part of mechanical energy (Jin et al., 2016). The
distribution along the outer surface of the spool head. appearance of cavitation bubbles also changes the posi-
Cavitation intensity and length increase with increasing tion of the actual minimum effective flow of the cross-
Pin . The maximum vapor volume fraction is nearly 0.95 at section and decreases its effective flow area. The shrink-
Pin = 4.0 MPa, and the maximum value is approximately ing cross-section area increases the pressure loss and
0.30 at Pin = 2.2 MPa. During the initial cavitation stage, reduces the regulating capacity of the valve. Therefore,
308 X. LIU ET AL.

system is set up to capture cavitation images, and the

digital image-processing techniques based on gray-level
detection on the images are used to calculate the cavita-
tion length. A 3D numerical model is built to calculate
the cavitation characteristic inside the regulating valve.
The numerical and experimental results show that bub-
bles appear near the throat, grow along the inner surface
of the throat and form a bubble ring. The cavitation devel-
opment process is separated into no-cavitation, initial
cavitation and steady-cavitation stages. In the initial cav-
itation stage, the cavitation area and intensity increase
with increasing Pin in the radial and axial directions. Dur-
ing the steady-cavitation stage, the cavitation area and
intensity remain stable even if Pin increases further. The
increasing Pin also affects the outline of cavitation bub-
bles. A slit exists between the upper and lower bubbles
when Pin is small. With the continuous development of
Pin , the cavitation area enlarges, and the upper and lower
parts of bubbles touch each other and form as a whole.
The inlet pressure influences the growth rate of cavita-
tion intensity, maximum velocity and flow capacity in the
regulating valve.
Although beneficial results have been obtained in our
study, we only focus on the steady-state characteris-
tics of the cavitation development process of the regu-
lating valve. The regulating valve usually works under
high temperature and pressure; consequently, the cav-
itation flow in the valve is a challenging issue. We
need to optimize relative interpretative mechanisms in
the future.
Figure 12. Numerical results: (a) vapor volume fraction on inner
surface of valve seat and (b) vapor volume fraction on outer
surface of spool head in the throat. Acknowledgements
Xiumei Liu designed, supervised the project and revised the
manuscript. Zihong Wu performed experiments and wrote the
draft. Beibei Li analyzed and Jiyun Zhao discussed the data.
Chi Zhang helped to carry out experiments. Li Wei gave some
advices about interpreting data. Fangwei Xie gave some advice
about revision. All authors discussed the results, implications
and commented on the manuscript at all stages.

Disclosure statement
No potential conflict of interest was reported by the authors.

Figure 13. Numerical maximum velocity when the

Schnerr–Sauer cavitation model is either employed or not. Funding
This research was funded by National Natural Science Founda-
the cavitation produces a negative effect on the flow
tion of China [grant number 51875559, U1810123]; National
capacity of the regulating valve. Key R&D Program of China [grant number 2016YFC0802908];
Natural Science Foundation of Jiangsu Province [grant num-
6. Conclusion ber BK20171189]; Fundamental Research Funds for the Cen-
tral Universities of China [2015XKMS026] and the Priority
The cavitation development process under different inlet Academic Program Development of Jiangsu Higher Education
pressures is investigated in our study. An experimental Institutions (PAPD).
 ENGINEERING APPLICATIONS OF COMPUTATIONAL FLUID MECHANICS 309

References method for cavitation-induced primary break-up. Fuel, 233,

778–795.
Abdulaziz, A. M. (2014). Performance and image analysis of
Liu, C. B., Jiang, S. L., & Zheng, Y. G. (2012). Experimental and
a cavitating process in a small type venturi. Experimental
computational failure analysis of a valve in a nuclear power
Thermal and Fluid Science, 53, 40–48.
plant. Engineering Failure Analysis, 22, 1–10.
Akbarian, E., Najafi, B., Jafari, M., Faizollahzadeh Ardabili, S.,
Liu, B., Zhao, J., & Qian, J. (2017). Numerical analysis of
Shamshirband, S., & Chau, K.-W. (2018). Experimental and
cavitation erosion and particle erosion in butterfly valve.
computational fluid dynamics-based numerical simulation
Engineering Failure Analysis, 80, 312–324.
of using natural gas in a dual-fueled diesel engine. Engineer-
Long, X., Zhang, J., Wang, J., Xu, M., Lyu, Q., & Ji, B. (2017).
ing Applications of Computational Fluid Mechanics, 12(1),
Experimental investigation of the global cavitation dynamic
517–534.
behavior in a venturi tube with special emphasis on the cav-
Amirante, R., Distaso, E., & Tamburrano, P. (2014). Exper-
ity length variation. International Journal of Multiphase Flow,
imental and numerical analysis of cavitation in hydraulic
89, 290–298.
proportional directional valves. Energy Conversion and Man-
Margot, X., Hoyas, S., Gil, A., & Patouna, S. (2014). Numerical
agement, 87, 208–219.
modelling of cavitation: Validation and Parametric studies.
Andriotis, A., Gavaises, M., & Arcoumanis, C. (2008). Vortex
Engineering Applications of Computational Fluid Mechanics,
flow and cavitation in diesel injector nozzles. Journal of Fluid
6(1), 15–24.
Mechanics, 610, 195–215.
Mosavi, A., Shamshirband, S., Salwana, E., Chau, K.-W., & Tah,
Chern, M.-J., Wang, C.-C., & Ma, C.-H. (2007). Performance
J. H. M. (2019). Prediction of multi-inputs bubble column
test and flow visualization of ball valve. Experimental Ther-
reactor using a novel hybrid model of computational fluid
mal and Fluid Science, 31(6), 505–512.
dynamics and machine learning. Engineering Applications of
Cui, B., Lin, Z., Zhu, Z., Wang, H., & Ma, G. (2017). Influ-
Computational Fluid Mechanics, 13(1), 482–492.
ence of opening and closing process of ball valve on external
Mou, B., He, B.-J., Zhao, D.-X., & Chau, K.-W. (2017). Numer-
performance and internal flow characteristics. Experimental
ical simulation of the effects of building dimensional varia-
Thermal and Fluid Science, 80, 193–202.
tion on wind pressure distribution. Engineering Applications
Dai, X., Wang, Z., Liu, F., Wang, C., Sun, Q., & Xu, C. (2019).
of Computational Fluid Mechanics, 11(1), 293–309.
Simulation of throttling effect on cavitation for nozzle inter-
Qiu, T., Song, X., Lei, Y., Dai, H., Cao, C., Xu, H., & Feng, X.
nal flow. Fuel, 243, 277–287.
(2016a). Effect of back pressure on nozzle inner flow in fuel
de Graaf, K. L., Brandner, P. A., & Pearce, B. W. (2017). Spectral
injector. Fuel, 173, 79–89.
content of cloud cavitation about a sphere. Journal of Fluid
Qiu, T., Song, X., Lei, Y., Liu, X., An, X., & Lai, M. (2016b).
Mechanics, 812, R1-1–13. doi:10.1017/jfm.2016.819.
Influence of inlet pressure on cavitation flow in diesel nozzle.
Forder, A., Thew, M., & Harrison, D. (1998). A numerical inves-
Applied Thermal Engineering, 109, 364–372.
tigation of solid particle erosion experienced within oilfield
Qiu, T., Wang, K., Lei, Y., Wu, C., & Guo, P. (2018a). Inves-
control valves. Wear, 216(2), 184–193.
tigation on effects of upstream flow on submerged jet flow
Ghalandari, M., Mirzadeh Koohshahi, E., Mohamadian, F.,
from short cylindrical orifice in common-rail injector. Inter-
Shamshirband, S., & Chau, K. W. (2019). Numerical simula-
national Journal of Heat and Mass Transfer, 127, 16–20.
tion of nanofluid flow inside a root canal. Engineering Appli-
Qiu, T., Wang, K., Lei, Y., Wu, C., Liu, Y., Chen, X., & Guo, P.
cations of Computational Fluid Mechanics, 13(1), 254–264.
(2018b). Investigation on effects of back pressure on sub-
Han, M., Liu, Y., Wu, D., Zhao, X., & Tan, H. (2017). A
merged jet flow from short cylindrical orifice filled with
numerical investigation in characteristics of flow force under
diesel fuel. Energy, 162, 964–976.
cavitation state inside the water hydraulic poppet valves.
Ramezanizadeh, M., Alhuyi Nazari, M., Ahmadi, M. H., &
International Journal of Heat and Mass Transfer, 111, 1–16.
Chau, K.-W. (2019). Experimental and numerical analysis
Jin, Z., Chen, F., Qian, J., Zhang, M., Chen, L., Wang, F., & Fei,
of a nanofluidic thermosyphon heat exchanger. Engineering
Y. (2016). Numerical analysis of flow and temperature char-
Applications of Computational Fluid Mechanics, 13(1), 40–47.
acteristics in a high multi-stage pressure reducing valve for
Soyama, H., & Hoshino, J. (2016). Enhancing the aggressive
hydrogen refueling station. International Journal of Hydro-
intensity of hydrodynamic cavitation through a venturi tube
gen Energy, 41(12), 5559–5570.
by increasing the pressure in the region where the bubbles
Jin, Z. Y., Gao, Z. X., Zhang, M., & Qian, J. Y. (2017). Pres-
collapse. Aip Advances, 6(4), 045113.
sure drop analysis of pilot-control globe valve with differ-
Wang, J., Xu, S., Cheng, H., Ji, B., Zhang, J., & Long, X. (2018).
ent structural parameters. Journal Of Fluids Engineering-
Experimental investigation of cavity length pulsation charac-
Transactions Of the Asme, 139(9), 12.
teristics of jet pumps during limited operation stage. Energy,
Jin, H., Zheng, Z., Ou, G., Zhang, L., Rao, J., Shu, G., & Wang, C.
163, 61–73.
(2015). Failure analysis of a high pressure differential regu-
Wei, M., Gao, Y., Yan, F., Chen, L., Feng, L., Li, G., & Zhang,
lating valve in coal liquefaction. Engineering Failure Analysis,
C. (2017). Experimental study of cavitation formation and
55, 115–130.
primary breakup for a biodiesel surrogate fuel (methyl
Laı´n, S., Bröder, D., Sommerfeld, M., & Göz, M. F. (2002).
butanoate) using transparent nozzle. Fuel, 203, 690–699.
Modelling hydrodynamics and turbulence in a bubble col-
Yang, H., Wang, W., & Lu, K. (2019). Numerical Simulations
umn using the Euler–lagrange procedure. International Jour-
on flow characteristics of a Nozzle-Flapper Servo valve With
nal of Multiphase Flow, 28(8), 1381–1407.
Diamond Nozzles. Ieee Access, 7, 28001–28010.
Li, D., Liu, S., Wei, Y., Liang, R., & Tang, Y. (2018). Numerical
Yin, B., Yu, S., Jia, H., & Yu, J. (2016). Numerical research of
investigation on transient internal cavitating flow and spray
diesel spray and atomization coupled cavitation by Large
characteristics in a single-hole diesel injector nozzle: A 3D
310 X. LIU ET AL.

Eddy simulation (LES) under high injection pressure. Inter- oil slurry valve. In B. Li, R. Baliga, & S. Finneran (Eds.),
national Journal of Heat and Fluid Flow, 59, 1–9. Proceedings of the asme pressure vessels and piping conference
Yuan, C., Song, J. C., & Liu, M. H. (2019). Comparison of (Vol. 3A, pp. 65237-1–7). Waikoloa, Hawaii, USA: Amer Soc
compressible and incompressible numerical methods in sim- Mechanical Engineers, Three Park Avenue, New York, NY
ulation of a cavitating jet through a poppet valve. Engineer- 10016-5990 USA.
ing Applications Of Computational Fluid Mechanics, 13(1), Zheng, Z., Ou, G., Ye, H., Zhang, L., Wang, C., Jin, H., & Shu,
67–90. G. (2016). Investigation on failure process and structural
Zhang, J., Mao, J.-R., Wang, S.-S., Wu, B., Yuan, H., Wang, optimization of a high pressure letdown valve. Engineering
K.-G., . . . Sun, P. (2014). Experimental study on the cavita- Failure Analysis, 66, 223–239.
tion characteristics of an oil jet pump with multiple nozzles Zheng, Z., Ou, G., Yi, Y., Shu, G., Jin, H., Wang, C., & Ye,
for the lubrication system in a Steam turbine. Journal of H. (2016). A combined numerical-experiment investigation
Engineering for Gas Turbines and Power, 136(12), 122605. on the failure of a pressure relief valve in coal liquefaction.
Zheng, Z., Ou, G. F., & Jin, H. Z. (2017). Numerical- Engineering Failure Analysis, 60, 326–340.
experimental study on the erosion-cavitation wear of coal