Hindawi

Geofluids
Volume 2020, Article ID 8847087, 12 pages
https://doi.org/10.1155/2020/8847087

Research Article
A Prediction Method of Wear for Volute Casing of a Centrifugal
Slurry Pump

Baocheng Shi , Jianpeng Pan, Lijuan Wu, Xingkai Zhang, Yijie Qiu, and Yindi Zhang
School of Petroleum Engineering, Yangtze University, Wuhan 430100, China

Correspondence should be addressed to Baocheng Shi; shibaoch@yangtzeu.edu.cn

Received 11 August 2020; Revised 18 September 2020; Accepted 19 November 2020; Published 18 December 2020

Academic Editor: Jinze Xu

Copyright © 2020 Baocheng Shi et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Volute wall wear situations directly affect a long time safe operation for the centrifugal slurry pump unit and the whole system. In
the present study, internal flow field is numerically investigated in a solid-liquid centrifugal pump, and the volute wall wear caused
by the solid-liquid two-phase flow is predicted with wear equation. A systematic analysis on the wear mechanism of the centrifugal
pump volute wall is carried out deeply, including the volute wall wear region, wear rate, and the relationship among inlet flow rate,
particle concentration, and particle size. The predicted high erosion intensity area shows good agreement with the experimental
erosion area, and the predicted and experimental areas are both located at the volute angle of 180° and near tongue. Therefore,
the wear equation put forward in the present study is effective for estimating the erosion intensity and predicting the erosion
area around the volute casing of a centrifugal pump.

1. Introduction particle concentration, and particle size. At present, most

results are only from scene experience. However, volute wall
When the two-phase centrifugal pump runs in a high speed wear conditions are very important, which influence the
rotation, the wear of the impeller and volute has become working state and the using of longevity of the centrifugal
one of the important reasons causing the material damage pump unit and the whole system. In the present study, inter-
or equipment failure [1, 2]. The continuing impact force nal flow field was numerically investigated in a solid-liquid
and the friction, which are caused by interparticle collisions centrifugal pump, and the volute wear caused by the solid-
and collisions between solid particles and impeller and liquid two-phase flow was predicted with wear equation.
volute, seriously affect the service life of pump, and the We analyzed the wall wear mechanism for different parts of
impeller and volute are the most vulnerable parts of wear volute by considering the influences of structural parameters,
and tear. The present study is mainly focused on developing operation parameters, and particle properties on the volute
a quantitative prediction method to investigate the wear of wall wear.
volute through numerical analysis. As we know, the research Currently, in order to effectively control and reduce the
in the influences of the liquid-solid two-phase flow on volute loss caused by wear and to improve the service life of equip-
wall wear in the solid-liquid two-phase centrifugal pump is ment and materials, researchers have extensively investigated
still in their infancy due to the complex phenomena and fac- solid-liquid two-phase flow rules inside the pump and wall
tors of erosion wear. These factors interact with one another, wear mechanism caused by solid particles. Most of the stud-
and the change of one factor may create a chain reaction in ies are centered on the analysis of particle properties and tra-
the volute wall wear rate, resulting in the change of wear jectories by experiments and calculations during the process
mechanism. There have been short of systematic analysis of wall collision wear under different flow parameters. Based
and comprehensive research on the wear mechanism of the on the analysis of boiler pipeline failure, Tilly [3] found that
centrifugal pump volute, including the volute wall wear about a third wear was caused by abrasion. Liu [4] investigated
region, wear rate, and the relationship among inlet velocity, numerically silt abrasive erosion of hydraulic machinery for
2 Geofluids

low sand-water concentrations (mean sand concentration by In the present study, a solid-liquid two-phase centrifugal
volume (C v < 1%)). In his study, the turbulent flow velocity pump is taken as the research object. Based on CFD (compu-
fields, pressure distributions, and the trajectories of sand par- tational fluid dynamics), internal flow field is calculated, and
ticles in the hydraulic machinery flow were numerically cal- the wear equation is employed to calculate the volute wear
culated using the mixed Eulerian-Lagrangian turbulence and wear area caused by the solid-liquid two-phase flow.
model. In addition, the sand particle rebound model and On the basis of the above analysis, the wear mechanism of
the erosion wear model in the hydraulic machinery flow field different parts of the volute is analyzed. The influences of
were also presented. Li and Han [5] simulated the flow field structural parameters, operation parameters, and particle
of the slurry pump by using a k-ε two equations model and properties on the volute wall wear are analyzed, and the
SMPLE algorithm. The simulation results suggested that the volute wall wear mechanism is obtained. These studies pro-
concentration of the solid phase and the wear in the blade vide a theoretical and practical basis for the optimization
and the volute increased with the increase of particle diame- design and safe operation of the solid-liquid two-phase cen-
ter and initial concentration of the solid phase. However, trifugal pump.
when the particle diameter and initial concentration of the
solid phase were increased to a certain value, the concentra- 2. Numerical Methods
tion of the solid phase in the volute would keep unchanged.
By comparison analysis, Shu, Xu, and Tang [6] found that 2.1. Governing Equations. For the two-phase flow centrifugal
small particles were liable to impact the rear-end surface of slurry pump at low solid phase volume fractions (usually no
impeller adjacent to the outlet side, while large solid particles more than 10%), the solid-liquid two-phase flow can be
could repeatedly impact the machine surface more easily numerically simulated with the discrete phase model. The
than small particles. The numerical simulation results were characteristics of the discrete phase model (DPM) are one-
verified by experimental results. An extensive parametric way interaction. The fluid, as a carrier, has influences on
study has been carried out in order to optimize the shroud- the movement of the particles by drag, drop, and vortex,
type impeller taking into account the blade discharge angle, while the solid particle effects on the liquid phase flow which
addition splitter blades, and modified blade (backward long is very limited and can be negligible.
blades) and find that the hydraulic efficiency of the centrifu- As the continuum phase in the centrifugal slurry pump,
gal slurry pump can be increased up to 9% by using the back- water is incompressible, thus the Renault eddy viscosity
ward long blades in addition to modified volute compared to model is employed to calculate the liquid flow field. The con-
the original ones [7, 8]. tinuity equation, momentum equation, and RNG (renorma-
Because of the complexity of the material erosion behav- lization group) k-ε turbulence model equations can be
ior, there is no acceptable, universally applicable formula to referred to the literatures [20–24].
predict erosion resistance of materials. So far, no reasonable In numerical calculations, the solid particle movement
explanation has been given for the material size effect or can be described by Lagrange’s equation. Based on a single
brittle-ductile transition, and the influences of particle trajec- particle forces analysis, the particle trajectory can be solved
tories on the material erosion behavior and particle size dis- by integral of differential equations. According to Newton’s
tribution on the plastic material erosion behavior are not second law, control equations for spherical particles in the
clearly reported. Since the first erosion theory, micro cutting slurry pump flow field in the x direction in the absolute coor-
theory, several similar theories have been put forward to dinates system can be given as follows:
explain or predict the erosion behavior of materials [9–16].
Unfortunately, none of these theories can successfully reveal 1 3   dup
πd ρ F + F p + F d + F v = mp , ð1Þ
the inner mechanism of the material erosion. Tuzson [17] 6 p p b dt
found that erosion damage accumulated over a considerable
time period, and the material removal rate was found to be where F b is the buoyancy, F b = ðρp − ρl /ρp Þgxi ; F p is the
approximately proportional to the local rate of power dissi- resistance caused by flow pressure gradient, F p = ðρl /ρp Þ
pation. The erosion depth would then be proportional to up ð∂u/∂xi Þ; F d is the drag, F d = ð18μ/ρp d2p ÞðC d Rep /24Þðul
the local energy dissipated by the slurry particles. Wilson, − up Þ; F v is the additional mass force, F v = 1/2ðρl /ρp Þðdðul
Addie, and Sellgren [18] believed that the removal of material
over a time occurs through small-scale deformation, cutting, − up Þ/dtÞ; dp , ρp , mp and up are the solid particle diameter,
fatigue cracking, or a combination of these and thus depends density, mass, and velocity, respectively, ρl and ul are the den-
on the properties of both the wearing surface and the parti- sity and velocity of liquid; gxi is the gravitational acceleration;
cles. Ductile materials tend to exhibit erosion primarily by μ is the dynamic viscosity of the liquid; C d is the drag coeffi-
deformation and cutting, with the specific type depending cient; and Rep is the particle Reynolds number.
on the angularity of the eroding particles. Brittle or hardened In the present study, the slurry flow through the centrif-
materials tend to exhibit fatigue cracking erosion under ugal pump was modeled using the DPM approach with a
repeated particles impact. The useful life of the most slurry multiphase model in FLUENT®. The RNG turbulence model
pump is limited by erosive wear of the flow passages. Sund- with swirling flow correction was employed in the turbulent
strom and Rendon [19] pointed out that particles with lower core region, and the wall function method was selected to
hardness than that of the abraded surface results in much treat the low Re flow near the solid wall. The SIMPLE algo-
lower wear rates than harder particles. rithm was applied to deal with pressure-velocity coupling.
Geofluids 3

Table 1: The basic design parameters of the impeller.

Rated flow Design head Specific speed Vane number Impeller outer diameter Impeller inlet diameter Blade inlet/outlet incidence
Q0 = 15:1L/s H = 15:1m ns = 85 Z=5 D2 = 228mm D0 = 100mm β1 = 15∘ , β2 = 30∘

Lateral clearance of
impeller cover

Inlet passage

(0,0,0)

Volute
casing

Figure 1: The whole flow field model (Shi et al. [21]).

 k2  k3
Second order upwind difference scheme was used for the dis- Vp Dp
cretization of momentum, k, and ε terms. E90 = K ðH v Þk1 , ð4Þ
V′ D′
2.2. Prediction Model of Wear. Erosion models can be used to
calculate the damage caused by particle impact on the wall.
These models include Finnie [9], Oka [14, 15], and McLaury n1 = s1ðH v Þq1 , n2 = s2ðH v Þq2 , k2 = 2:3ðH v Þ0:038 : ð5Þ
[25]. The default constant of the Finnie erosion model is only
applicable to the erosion calculation of the carbon steel wall
impacted by sand particles. McLaury [25] proposed a model Here, EðθÞ is the wear loss in an arbitrary impact
for predicting the erosion rate of sand in water, which is angle(mmkg/s), E90 is the wear loss in a 90° impact
mainly used to simulate the erosion rate in the process of angle(mmkg/s), H v is the Vickers hardness(Gpa), V p and
mud erosion. The model constants and both need to be deter- V ′ are the particle impact velocity and the reference impact
mined by experiments. The model constants given in the lit-
velocity(m/s), and Dp and D ′ are the particle diameter and
erature are only valid for water and sediment suspensions.
The particle collision velocity is assumed to be in the range the reference particle diameter (μm).Other parameters are
of 0-10 m/s. In actual flow, the solid particle velocity and col- given as
lision angle are widely distributed, and the simulated parti-
cles come in a variety of shapes and sizes, so an ideal wear s1 = 0:71, s2 = 2:4, q1 = 0:14, q2 = −0:94, K = 65, k1 = −0:12, k3 = 0:19:
equation should be applied to different simulated calculation
very well and should not have too many restrictions. The Oka ð6Þ
wear equation [14, 15] is applicable to any hardness material,
impact angle, and impact speed with a broader scope and rel- The wear rate is defined by
atively fewer restrictions than other wear equations, so it is
selected in the present study. The Oka wear equation can

N paticles
be given as follows: _ p C dp f ðaÞvbðvÞ
m
Rerosion = 〠 : ð7Þ
p=1 Aface
EðθÞ = gðθÞE90 , ð2Þ

Here, Cðd p Þ is a function of d p , f(a) is a function of

gðθÞ = ðsin θÞn1 ½1 + H v ð1 − sin θÞn2 , ð3Þ the impact angle a, and v is the impact velocity function.
4 Geofluids

Volute-type
casing
Discharge
Z X
channel

125 cm
100 cm

Inlet channel

Figure 2: Mesh (Shi et al. [21]).

20
72
18
64

16 56

14 48

𝜂 (%)
H (m)

12 40

32
10
24
8
16
6
8
4
0 2 4 6 8 10 12 14 16
Q (m3/s)

Test of clear water Test calculated values of clear water

Test of glass beans Test calculated values of glass beans

Figure 3: Comparison of the calculated performances with the experimental values.

Substitute the above wear equations (2)–(6) into wear rate Considering the unit size, a new wear rate equation
equation (7): can be given:

 k3 N paticles
_ p EðαÞ10−6
m

Dp Rerosion = 〠 : ð9Þ
C dp = ; Aface
D′ p=1

f ðαÞ = ðsin αÞn1 ½1 + H v ð1 − sin αÞn2 E90 K ðH v Þk1 ; ð8Þ

Normal coefficients can be given:
 k2
bðvÞ
Vp
v = :
V′ εN = 0:993 − 0:0307α + 4:75 × 10−4 α2 − 2:61 × 10−6 α3 : ð10Þ
Geofluids 5

Pressure
0.15
6.2E+04 6.2E+04

6.0E+04
0.1
5.5E+04 6.0E+04 4.5
E+
04

0 4
E+
5.0E+04

2.5
0.05

Y (m)
4.5E+04

4.0E+04
6.0E+04 0
4.5E+04
3.5E+04
4.5E+04
3.0E+04 −0.05
2.5E+04 5.5E+04
6.0E+04
2.0E+04 −0.1

−0.1 −0.05 0 0.05 0.1

X (m)
(a) Static pressure distribution for the liquid phase in center section (Pa)
Velocity magnitude dpm-erosion rate
0.15
1.0E+01 4.0E+00 1.0E–03
9.0E–04
9.0E+00
0.1 8.0E–04
8.0E+00 7.0E+00 7.0E–04
00 6.0E–04
7.0E+00 E+
8.0
0.05 5.0E–04
6.0E+00 4.0E–04
Y (m)

6.0E+00 3.0E–04
5.0E+00 0 2.0E–04
4.0E+00 1.0E–04
7.0E+00
6.0E+00 6.8E–05
3.0E+00 −0.05
1.8E–05
2.0E+00 4.5E–06
6.0E+00
1.0E+00 −0.1 3.7E–06
1.8E–06
0.0E+00
−0.1 −0.05 0 0.05 0.1
X (m)
(b) Relative velocity distribution for the liquid phase in center section (m/s) (c) Dpm-erosion rate distribution along the volute wall (kgm-2s-1)

Figure 4: Continued.
6 Geofluids

Wall shear
2.2E+02

2.0E+02
1.8E+02
1.6E+02

1.4E+02
1.2E+02

1.0E+02
8.0E+01

6.0E+01
4.0E+01

2.0E+01

(d) The wall shear distribution along the volute wall (Pa)

Figure 4: Several important parameter distributions in the volute.

Tangent coefficients can be given:

εT = 0:998 − 0:029α + 6:43 × 10−4 α2 − 3:56 × 10−6 α3 : ð11Þ Flow

2.3. Pump Specifications and Calculation Conditions. We

applied the above models to a centrifugal pump. The impeller
of the pump has five blades and a volute-type casing, which is
made of zinc. The maximum size of the impeller is 160 mm,
and the hub diameter is 56 mm. Operating at the design point,
the flow rate of the pump, Q, is 15m3/h, the head, H, is 20 m, Direction
and the rotational speed is 1450 rpm. The main parameters of rotation of
the impeller are listed in Table 1. The different flow channels impeller
(balance hole, lateral clearances of impeller cover and disk,
impeller) of the pump model are shown in Figure 1. Figure 2 𝜃 = Volute angle from
shows a schematic of the numerical mesh and boundary con- the volute tongue in
ditions used in the simulation. Figures 1 and 2 are reproduced degree
from Baocheng and Shi et al., Numerical Simulation of 3D
Solid-Liquid Turbulent Flow in a Low Specific Speed Centrif- Figure 5: Schematic view of the volute casing with the location of
ugal Pump: Flow Field Analysis. 2015. A cylindrical suction volute angle.
channel with an axial length of 100 mm is connected to the
region including the impeller blades, and a 125 mm cylindri-
cal discharge channel is also connected downstream of the siequilibrium status. The time step in the second stage of the
volute-type casing. The total number of grids of the volute- simulation is 0.02 microseconds.
type casing including the inlet channel, and the discharge
channel is 812944. 3. Results and Discussion
The liquid velocity at the inlet boundary keeps constant of 3.1. Pump Performance. The slurry pressure value at the solid-
Q/A, m/s. The solid particle radius is in the range of 30- liquid pump inlet section (i.e., inhalation tube) and total outlet
300 μm at the inlet boundary. Nonslip conditions are assumed pressure value at the water pressure chamber can be obtained
for the solid surface of the blade and shroud of the volute, and by surface integral functions in the FLUENT ®. The pump
the assumption of nonslip conditions is also adopted for the head is defined as the total head difference between the inlet
wall surface of the suction and discharge channels. and outlet of the pump and can be expressed as
To reduce the calculation time, the simulation process is
divided into two stages and carried out sequentially. First, the
Pmout − Pmin
flow field is formed using local time stepping [25]. The solid Hm = + Δz, ð12Þ
particles motion in this stage is assumed to remain in a qua- ρm g
siequilibrium status to avoid instability in calculation when
the liquid phase pressure changes. Then, the solid-particle where H m is the slurry pump head of the solid-liquid pump,
behavior can be calculated in detail, without assuming a qua- m, Pmout is the total pressure at the outlet section of the slurry
Geofluids 7

Wear mass loss (kgm−2 s−1)

0.0
−30 0 30 60 90 120 150 180 210 240 270 300 330 360
Volute angle (degree)

15 m3/h
10 m3/h

Figure 6: Variation in wear along the outer casing wall at two flow rates for the solid-liquid mixture of 5%concentration (by volume).
Wear mass loss (kgm−2 s−1)

0.0
−30 0 30 60 90 120 150 180 210 240 270 300 330 360
Volute angle (degree)

5%
10%

Figure 7: Variation in wear along the outer casing wall at two solid concentrations for the solid-liquid mixture at 15m3/h discharge rate.

solid-liquid pump, Pa, Pmin is the total pressure at the inlet sec- flow field, a reasonable efficiency prediction formula is put for-
tion of slurry solid-liquid pump, Pa, and Δz is the position ward to calculate the efficiency of solid-liquid pump as follows:
head, namely, the difference of centre levels between the pump
outlet passage and inlet passage in the vertical direction, m. ρm gQH m
ηp = × 100%, ð13Þ
Pmout and Pmin can be obtained by the surface integral function Mω
in FLUENT software, and slurry density ρm and acceleration
of gravity g are known parameters, so the head of the solid- where ηp is the efficiency of the two-phase flow pump, Q is the
liquid pump can be achieved conveniently and accurately by volume flow rate of the two-phase flow m3/s, M is the impeller
Eq.(12). torque, N ⋅ m, and ω is the impeller angular velocity, rad/s.
In order to improve the traditional efficiency prediction Slurry density ρm , acceleration of gravity g, and impeller angu-
methods, based on the 3D numerical simulation of the whole lar velocity ω are known parameters, H m can be obtained by
8 Geofluids

Wear mass loss (kgm−2 s−1)

0.0
0 2 4 6 8 10 12 14 16 18 20
Q (m3/h)
The maximum values
The average values

Figure 8: Variation of wear mass loss of volute material with the inlet flow rate.

180

160

140

120
The wall shear (Pa)

100

80

60

40

20

0
0 2 4 6 8 10 12 14 16 18 20
Q (m /h)
3

The maximum values

The average values

Figure 9: Variation of wall shear with the inlet flow rate.

Eq.(12), and impeller torque M can be obtained from a forces the experimental data, with the head error less than 2% and
report in FLUENT®. efficiency error 4% at the design point. The results of com-
The comparisons of head and efficiency between the cal- parison indicate that the method selected in the present study
culated results and experimental data under the rotational has good performance and reliability for particle-fluid two-
speed of 1450 rpm are shown in Figure 3. From the compar- phase flows in the centrifugal slurry pump.
ison results of clear water and the two-phase flow with the
solid-phase volumetric concentration of 10% and particle 3.2. Static Pressure, Velocity, Dpm-Erosion Rate, and Wall
radius of 0.002 mm, we find that the total head was equiva- Shear. Figure 4 shows the predictions for instantaneous dis-
lent to the pressure increment between the outlet of the suc- tributions of static pressure, velocity for the liquid phase in
tion pipe and the outlet of the volute, and the calculated the center section, the dpm-erosion rate, and the wear shear
results for clean water and the two-phase flow are close to along the volute wall. The pressure of the fluid shows the
Geofluids 9

Wear mass loss (kgm−2 s−1)

0.0

−50 0 50 100 150 200 250 300 350

Particle diameter (𝜇m)

The maximum values

The average values

Figure 10: Variation of wear mass loss of volute material with the particle diameter.

150

120
The wall shear (Pa)

90

60

30

0
0 50 100 150 200 250 300 350

Particle diameter (𝜇m)

The maximum values
The average values

Figure 11: Variation of wall shear with the particle diameter.

trend of increasing from the impeller out to the volute wall, maximum wall shear and erosion rate are both in the vicinity
and the low-pressure region is formed near the tongue owing of the volute tongue and volute throat at θ = 180° . This indi-
to the higher velocity here, as shown in Figure 4(a). cates that the impact angle of the solid particles decreases
Figure 4(b) presents the velocity variations inside the volute with increasing volute angle. Hence, it can be concluded that
for the liquid phase. The velocity distribution appears to be the variation in the impact angle of the solid particles is pri-
nonuniform, and the maximum values occur near the ton- marily responsible for the wear pattern observed at various
gue. Two higher velocities occur at the volute angle of 30° operating conditions of the pump. But in the zone near to
and 180°, respectively. Comparison of Figure 4(c) with the volute tongue and volute throat, it is not feasible to iden-
Figure 4(d) shows that the dpm-erosion rate and the wall tify the angle of impact, and these phenomena are also
shear distributions along the volute wall are similar, and the observed in the numerical prediction of trajectory for solid
10 Geofluids

Wear mass loss (kgm−2 s−1)

0.0

−2 0 2 4 6 8 10 12
Particle concentration (%)
The average values
The maximum values

Figure 12: Variation of wear mass loss of volute material with particle concentration.

150

120
The wall shear (Pa)

90

60

30

0
−2 0 2 4 6 8 10 12
Particle concentration (%)

The average values

The maximum values

Figure 13: Variation of wall shear with particle concentration.

particles in a concentric-type casing by Minemura and increase of in the volute angle and attains the maximum
Zhong [26]. value at θ = 180° . The maximum wear around similar loca-
tion for coatings of polyamide and epoxy resins on different
3.3. Variation of Dpm-Erosion Rate with the Volute Angle. casings with varying configurations (annular, semiannular,
The variation of the dpm-erosion rate with the volute angle and quasispiral) was also observed in the prediction of Roco
(as defined in Figure 5) is presented graphically in Figure 6 et al. [27], when operating the pump at the BEP flow rate.
for a solid concentration of 5% (by volume) at two discharge With further increase in the volute angle, the dpm-erosion
rates, namely, 10 m3/h and 15 m3/h. The discharge rate of 15 rate decreases, and the rate of decrease is higher for angles
m3/h is the BEP (best efficiency point) flow rate for the beyond 180°. Similar observations based on the measurement
assembled pump. At this discharge rate, we find that the of loss in wall thickness have been reported by Wiedenroth
dpm-erosion rate decreases first and then increases with the [28] for highly wear-resistant volute casing and by Roco
Geofluids 11

et al. [27] for coatings of resin material at the volute surface. fields. For proper design and selection of a slurry pump,
We also observed that the wear mass loss at the flow rate of extensive data are required for accurate estimation of the
10m3/h is a little higher than that at the BEP flow rate for pump performance for the case of the slurry flow at high con-
the assembled pump. At lower flow rates, the flow velocities centrations. Based on the present investigations, it is con-
may be a little higher for volute angles less than 320° due to cluded that the relative wear along the casing is a function
higher recirculating flow in the volute, resulting in slightly of the erosion behavior of the volute casing material and var-
higher wear at 10 m3/h discharge rate than that at 15 m3/h iation of impact angles of solid particle around the casing. It
(105% of BEP flow rate). is seen that the erosion is less at flow rates close to BEP than
Figure 7 shows the trend of simulation curves of the that at lower flow rates. The high erosion intensity areas pre-
dpm-erosion rate with the volute angle for 5% and 10% solid dicted by the method are distributed in the vicinity of a volute
concentration at 15m3/h discharge rate. As shown in the fig- angle of around 180°and near the volute tongue. The results
ure, the variation and shape of the curve observed at 10% can provide a theoretical and practical reference for the
solid concentration appears similar to the one at 5%. Com- two-phase flow pump optimization design to get a long
pared to Figure 7, variation in wear along the outer casing period of safe operation.
wall at concentration of 5% and 10% is also similar to that
at 8% concentration for the same discharge rate. Moreover, Data Availability
the dpm-erosion rate for 10% solid concentration is 10%
-40% higher than the corresponding values for 5% at the The data used to support the findings of this study are
same discharge rate. included within the article.

3.4. Influence of Particle Parameters on Wear Mass Loss. Conflicts of Interest

Figure 8 shows the volute wear mass loss of material chang-
ing with the inlet flow rate. It can be seen that the maximum The author(s) declare(s) that they have no conflicts of
wear values and the average wear values increase with the interest.
increase of the inlet flow rate. This indicates that the increas-
ing velocity increases the probability of particle collision with Acknowledgments
the wall. Similar variation of wall shear with the inlet flow
rate can be seen from Figure 9. The authors gratefully expressed their thanks for the financial
Figure 10 shows that the maximum volute wear mass loss support by the National Natural Science Foundation of
of material and the average wear mass loss of material decrease China (No. 51974033) and Educational Commission of
with the increase of particle diameter. The maximum wear Hubei Province of China (D20171305).
mass loss of material curves shows an obvious downward
trend with the particle diameter when the particle diameters References
are less than 150 μm. This result indicates that the increasing
solid particle diameter reduces the particle collision frequency [1] G. Dong and J. Zhang, “Developments of research on the solid
with wall, and the particle collision energy is not the only rea- particle erosion of materials,” Journal of Materials Science and
son for the mass loss of volute material. The variations of wall Engineering, vol. 21, no. 2, pp. 307–312, 2003.
shear with the particle diameter and particle concentration are [2] Y. Ma, J. Ren, and Y. Li, “Development of research on erosion
shown in Figures 11 and 12, respectively. The wall shear of materials,” Journal of Lanzhou University of Technology,
changes a little with variation of the particle diameter and par- vol. 31, no. 1, pp. 21–25, 2005.
ticle concentration. Therefore, it can be included that the two [3] G. P. Tilly, “Erosion caused by impact of solid particles,” in
parameters have limited influence on the wall shear. Wear, vol. 13 of Treatise on Materials Science & Technology,
The variation of wear mass loss of volute material with , pp. 287–319, Elsevier, 1979.
particle concentration is shown in Figure 13. It can be seen [4] X. B. Liu, “Numerical simulation of silt abrasive erosion in
from the figure that the maximum wear values and the average hydraulic machinery,” Journal of Sichuan University of Science
wear values are both increasing with the increase of particle and Technology, vol. 19, no. 2, pp. 79–84, 2000.
concentration. This indicates that the increasing velocity [5] R. Li, Z. Xu, and W. Han, “Numerical simulation of interior
flow and wear character in slurry pumps,” Drainage and Irri-
increases the probability of particle collision with the wall.
gation Machinery, vol. 2, no. 25, pp. 5–11, 2007.
The maximum wear curve is in an increasing trend. The curve
[6] L. J. Shu, H. Y. Xu, and L. L. Tang, “Numerical simulation of
changes a little when the concentration is less than 6%, while
erosion and particle motion trajectory in centrifugal pump,”
the change is relatively large with further increasing the con- Transactions of the Chinese Society for Agricultural Machinery,
centration. This indicates that the increasing particle concen- vol. 6, 2008.
tration increases the probability of particle collision with the [7] M. S. Cellek and T. Engin, “3-D numerical investigation and
wall and enhances the erosion and abrasion of the volute wall. optimization of centrifugal slurry pump using computational
fluid dynamics,” Thermal Science and Technology, vol. 36,
4. Conclusions no. 1, pp. 69–83, 2016.
[8] M. S. Cellek and T. Engin, “Parametric investigation of a cen-
Slurry pumps are extensively used in the hydraulic transpor- trifugal slurry pump while handling clear water,” Journal of
tation of the solid-liquid mixture through pipes in various Thermal Science & Technology, vol. 36, no. 2, 2016.
12 Geofluids

[9] I. Finnie, “The mechanism of erosion of ductile metals,” 3rd [27] M. C. Roco, G. R. Addie, J. Dennis, and P. Nair, “Modeling
US National Congress of Applied Mechanics, vol. 25, pp. 527– erosion wear in centrifugal slurry pumps,” Hydrotransport,
532, 1958. vol. 9, 1984.
[10] J. G. A. Bitter, “A study of erosion phenomena part I,” Wear, [28] W. Wiedenroth, “Evaluation of wear distribution of a dis-
vol. 6, no. 1, pp. 5–21, 1963. mountable impeller in a model dredge-pump,” Fuel and
[11] X. Zhang, Z. Jiang, R. Liao et al., “Study on temperature distri- Energy Abstracts, vol. 5, no. 38, pp. 358–364, 1997.
bution of perforated horizontal wellbore,” Journal of Thermal
Science, vol. 29, no. 1, pp. 194–205, 2020.
[12] A. V. Levy, “The solid particle erosion behavior of steel as a
function of microstructure,” Wear, vol. 68, no. 3, pp. 269–
287, 1981.
[13] I. M. Hutchings, “A model for the erosion of metals by spher-
ical particles at normal incidence,” Wear, vol. 70, no. 3,
pp. 269–281, 1981.
[14] Y. I. Oka, K. Okamura, and T. Yoshida, “Practical estimation
of erosion damage caused by solid particle impact,” Wear,
vol. 259, no. 1-6, pp. 95–101, 2005.
[15] Y. I. Oka and T. Yoshida, “Practical estimation of erosion
damage caused by solid particle impact,” Wear, vol. 259,
no. 1-6, pp. 102–109, 2005.
[16] P. Buffiere and R. Moletta, “Collision frequency and collisional
particle pressure in three-phase fluidized beds,” Chemical
Engineering Science, vol. 55, no. 22, pp. 5555–5563, 2000.
[17] J. Tuzson, Centrifugal Pump Design, Wiley, New York, NY,
2000.
[18] K. C. Wilson, G. R. Addie, and A. Sellgren, Slurry Transport
Using Centrifugal Pumps, Springer Science & Business Media,
2006.
[19] A. Sundstrom, J. Rendon, and M. Olsson, “Wear behaviour of
some low alloyed steels under combined impact/abrasion con-
tact conditions,” Wear, vol. 250, no. 1-12, pp. 744–754, 2001.
[20] B. Wu, H. Yan, and J. Zhang, “Numerical simulation and per-
formance study on 3-D turbulence forecast of slurry pump,”
China Mechanical Engineering, vol. 20, no. 5, pp. 585–589,
2009.
[21] B. Shi and J. Wei, “Numerical simulation of 3D solid-liquid
turbulent flow in a low specific speed centrifugal pump: flow
field analysis,” Advances in Mechanical Engineering, vol. 6,
Article ID 814108, 2014.
[22] X. Zhang, D. Wang, R. Liao, S. Wang, B. Shi, and L. Wu,
“Associated petroleum gas measurement at low gas content
using PIS method,” Flow Measurement and Instrumentation,
vol. 70, p. 101662, 2019.
[23] X. Zhang, D. Wang, R. Liao, H. Zhao, and B. Shi, “Study of
mechanical choked Venturi nozzles used for liquid flow con-
trolling,” Flow Measurement and Instrumentation, vol. 65,
pp. 158–165, 2019.
[24] Q. Shen, Y. Jiang, F. Xia et al., “Hydrogen production by co-
based bimetallic nano-catalysts and their performance in
methane steam reforming,” Petroleum Science and Technology,
vol. 38, no. 6, pp. 618–625, 2020.
[25] B. S. McLaury, S. A. Shirazi, J. R. Shadley, and E. F. Rybicki,
“Modeling erosion in chokes,” in American Society of Mechan-
ical Engineers, Fluids Engineering Division (Publication) FED,
Vol. 236, pp. 773–781, ASME, 1996.
[26] K. Minemura, Y. Zhong, and T. Uchiyama, “Numerical Pre-
diction of Erosion Wear on Pump Casing under Solid-Water
Two-Phase Flow,” Multiphase Flow, vol. 61, pp. 561–572,
1995.