Numerical Simulation of Scouring Around Four Cylindrical Piles

with Different Inclination Angles Arrangements

Jyh-Haw Tang

and Aisyah Dwi Puspasari

Department of Civil Engineering, Chung Yuan Christian University, Zhongli District, Taoyuan City, Taiwan

Keywords: Numerical Simulation, Scouring, Pile Groups, Inclination Angles Arrangements, Flow-3D.

Abstract: One of the most frequent probable reasons for bridge pile foundation collapse, scouring can result in fatalities

as well as economic and environmental consequences. However, in the local scour simulation, there are few

thorough numerical investigations around pile groups consisting of front, middle, and rear piles with various

inclination degrees of flow attack. The maximum depth of local scour and its mechanism surrounding groups

of four cylinders with various arrangements of inclining angles were determined by numerical simulation

using Flow-3D software. A validation using the experimental investigation as a comparison was carried out

to verify the numerical model. Compared with experimental results, good conformity is shown in the

numerical results of scour depth and bed elevation contour using the Van Rijn transport rate equation and

RNG k-ε turbulence model. Numerical simulations of four cylinders in different alignment angles were carried

out with pile spacing ratios, G/D of 2.5, 3, and 3.5. The deepest maximum local scour always obtained at the

rear pile. The trend of maximum local scour depth is consistent with experimental studies, which

demonstrated the reliability and capability of the numerical model used to simulate pile groups in estimating

sediment scour depth.

1 INTRODUCTION

Scouring is the main cause of bridge failure due to the

reduced foundation resistance strength in the soil.

“Scouring is a natural phenomenon that occurs

because the presence of bridge pile foundation affects

the flow pattern and increases turbulence around the

pile foundation then removes and erodes the bed

material around the pile” (Ghaderi & Abbasi, 2019;

Jia et al., 2018; Zhang et al., 2017). The impact of

scouring around the pile foundation can cause

damage to the bridge structure, economic impact, and

loss of life (Tang & Puspasari, 2021). Therefore, the

study of scour becomes important during design,

operation, and maintenance (Kayser & Gabr, 2013;

Storey & Delatte, 2003), especially to monitor and

assess the scour depth obtained to reduce the adverse

impact on the hydraulic structure. There are three

methods used to estimate the scour depth, including

conventional, experimental, and numerical. The

difference is that the conventional or so-called

traditional method uses depth measuring instruments

https://orcid.org/0000-0003-2029-7799

https://orcid.org/0000-0002-1511-4057

depending on the high cost of installing and

maintaining the underwater monitoring, which has a

risk of damage during high floods (Lu et al., 2008),

whereas, in recent decades, experimental and

numerical methods have been used to study about

scouring around pile foundations. However, several

issues with the experimental method, such as limited

data output and inconsistent laboratory results

depending on environmental conditions, meaning the

results cannot be directly applied to guide design

practice or to field applications (Deng & Cai, 2010).

Hence, the numerical method is most suitable for use

in studying scour with low cost and time-efficient as

well as high accuracy and is effective for predicting

the probability of failure (Zhang et al., 2017).

Many researchers have studied the numerical

study of scour with various problems to comprehend

the scour mechanisms around pile foundations.

However, many numerical studies simulate the scour

depth around the pile group with the direction of

water flow in line with the pile position. In fact, there

are also conditions where the water flow is not in the

Tang, J. and Puspasari, A.

Numerical Simulation of Scouring Around Four Cylindrical Piles with Different Inclination Angles Arrangements.

DOI: 10.5220/0012115500003680

In Proceedings of the 4th International Conference on Advanced Engineering and Technology (ICATECH 2023), pages 139-145

ISBN: 978-989-758-663-7; ISSN: 2975-948X

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

139

same direction as the pile position but forms an

inclination angle. Several studies have discussed the

scour depth around the pile foundations with

inclination angle, including (Kim et al., 2017)

conducted a numerical study of two cylinders with

different alignment angles ranging from 0º – 90º

which resulted in the finding of increased scour depth

around the rear pile as the alignment angle increased

until it reached about 45º – 60º. (Zhang et al., 2017)

simulated a numerical study of sediment scour depth

around three cylindrical piles arranged at an angle of

45º from the direction of water flow with several pile

spacing ratios, G/D of 2, 3, and 4 using Flow-3D

software. This study found that the scour hole pattern

and G/D significantly impacted the contour of bed

elevation. Besides, the maximum scour depth

obtained increased from the front pile to the rear pile.

(Hamidi & Siadatmousavi, 2018) modeled the depth

of sediment scour around two cylinders in an

alignment arrangement with various angles of attack

(0, 15, 45, 60, and 90º) and G/D equal to 5 using

Sediment Simulation In Intakes with Multiblock

option (SSIIM) software. Compared to the

experimental results, the SSIIM model

underestimated the sediment scour depth obtained at

the front pile, while the good fit was at the rear pile.

Based on the literature review above, few studies

still discuss local scour simulation around pile groups

consisting of more than two piles with varying

inclination degrees of flow attack and pile spacing

ratio numerically. Meanwhile, a comprehensive study

including the scour mechanism is needed to

understand in determining the appropriate design. So,

numerical simulations of the scour depth around four

cylinders will be carried out with different inclination

angles arrangement and various pile spacing ratios

using Flow-3D software in this study. Further, to

confirm the applicability and accuracy of the

numerical model, a verification simulation was

validated by comparing the experimental result

before modeling the main case of the pile group.

2 MODEL VERIFICATION

A numerical model was set up similar to the

experimental study of (Khosronejad et al., 2012) to

validate applicability and accuracy using Flow-3D

software. “Flow-3D is a commercial Computational

Fluid Dynamics (CFD) software created by Flow

Science Inc., which can solve Navier-Stokes equation

for free-surface flow and solve complex meshing

geometries using Volume of Fluid (VOF) and

Fractional Area-Volume Obstacle Representation

(FAVOR) methods” (Flow Science, 2008).

Reynolds-Averaged Navier Stokes (RANS), which is

given below, is the equation that represents the three-

dimensional motion of a viscous fluid and governs the

motion of an incompressible flow around a bridge

pile foundation.

𝜕𝑢

𝜕𝑡

𝑉



𝑢𝐴



𝜕𝑢

𝜕𝑥

+ 𝑣𝐴



𝜕𝑢

𝜕𝑦

+ 𝑤𝐴



𝜕𝑢

𝜕𝑧



= −

𝜌

𝜕𝑝

𝜕𝑥

+ 𝐺



𝑓



(1)

𝜕𝑣

𝜕𝑡

𝑉



𝑢𝐴



𝜕𝑣

𝜕𝑥

+ 𝑣𝐴



𝜕𝑣

𝜕𝑦

+ 𝑤𝐴



𝜕𝑣

𝜕𝑧



= −

𝜌

𝜕𝑝

𝜕𝑦

+ 𝐺



𝑓



(2)

𝜕𝑤

𝜕𝑡

𝑉



𝑢𝐴



𝜕𝑤

𝜕𝑥

+ 𝑣𝐴



𝜕𝑤

𝜕𝑦

+ 𝑤𝐴



𝜕𝑤

𝜕𝑧



= −

𝜌

𝜕𝑝

𝜕𝑧

+ 𝐺



𝑓



(3)

where u, v, and w are the fluid velocity sections for

the x-, y-, and z-coordinates, V

is volume fraction,

is area fraction, ρ is fluid density, p is average

hydrodynamic pressure, and G

is body acceleration

and f

is viscous acceleration.

There are several parts of the model setup must be

considered in the Flow-3D software which is divided

into three stages including general, physics, then

meshing and geometry.

2.1 General Setup

In the general setup, the finish time and units are

defined according to the length of the simulation time

and the units used in the experiment. In this study, the

finish time was set to 3000 seconds with simulation

units of SI.

2.2 Physics Setup

In the physics setup, there are two important parts to

be assessed that will directly affect the simulation

results, namely the turbulence and sediment scour

models.

2.2.1 Turbulence Model

The renormalized group (RNG) k-ε turbulence model

was employed in this study utilizing the Flow-3D

software among several other models such as the

ICATECH 2023 - International Conference on Advanced Engineering and Technology

140

Prandtl mixing length model, the one-equation

model, the two-equation model (k-ε), the two-

equation model (k-ω) and the large eddy simulation

(LES) model because it more precisely and properly

characterizes low-intensity turbulent flow with the

stronger shear area (Zhang et al., 2017). Several

studies (Jalal & Hassan, 2020; Nazari-Sharabian et

al., 2020; Omara et al., 2018; Tang & Puspasari,

2021; Zhang et al., 2017) have proven by conducting

simulations to assess the accuracy of the RNG k-ε

turbulence model and produced a good agreement.

2.2.2 Sediment Scour Model

Several important parameters must be set to model

the sediment scour using Flow-3D software. First,

critical shields number θ

cr,i

corresponding to the

critical bed shear stress τ

required for removing

sediment from the packed bed interface (Brethour,

2003) was calculated using the Soulsby-Whitehouse

equation with a default value of 0.05 (Wei et al.,

2014). Second, bed load transport Ф

as the sediment

transport mode by rolling on the packed bed

sediment’s surface was calculated using the Van Rijn

equation with a bed-load coefficient value of 0.053

(Van Rijn, 1984). The maximum packing fraction has

a default value of 0.64 and the bed shear stress has a

recommended value of 2.5 (Wei et al., 2014). At last,

the sediment characteristics, including entrainment

coefficient and angle of repose have values of 0.005

(Omara et al., 2018) and 32º (Wei et al., 2014) for

sediments with a diameter size of an average particle

of 0.00085 m and the depth of 0.2 m (Khosronejad

et al., 2012).

2.3 Meshing and Geometry Setup

At this stage, the steps that must be completed are

modeling the geometry of channel flow then proceed

with adding the mesh and determining the boundary

conditions.

2.3.1 Meshing and Geometry

The geometry of the numerical simulation for

verification was modeled the same as the

experimental by (Khosronejad et al., 2012) with

channel dimensions 10 m long, 1.21 m wide, 4.5 m

high and 0.1651 m pile diameter D positioned 4 m

downstream from the inlet. A hydrodynamic entry

length L

of 6 m was added to develop the fluid

velocity before entering the sediment packed bed

which is modeled by solid red color in Figure 1 (a).

The flow depth and inflow velocity were 0.186 m and

0.25 m/s. To improve the precision of the model

surrounding the pile, 400,000 cells total, made up of

400 × 50 × 20 cells in the x, y, and z axes were

produced using mesh planes of the x and y directions

at a finer resolution as shown in Figure 1 (b). To

obtain the best geometry shape and surface model, all

meshing setups meet the requirements for maximum

adjacent and aspect ratios.

(a)

(b)

Figure 1: Numerical model for verification with the display

of (a) geometry and (b) meshing.

2.3.2 Boundary Conditions

The applied boundary conditions correspond to the

physical conditions in real life including the specified

velocity for the inlet, the wall for the front, rear, and

bottom sides, the specified pressure for the top

boundary, and the outflow for the outlet. Details of

boundary conditions is shown in Figure 2.

Figure 2: Boundary conditions.

Numerical Simulation of Scouring Around Four Cylindrical Piles with Different Inclination Angles Arrangements

141

2.4 Verification Result

The numerical simulation run for 50 minutes

produced the depth of sediment scour around the

single pile compared to experimental results, as

plotted in Figure 3. Moreover, the bed elevation

contours obtained from the numerical results

represent scour and deposition depths in various

colors compared to the experimental results.

Figure 3: Validation results of the sediment scour depth.

The greatest sediment scour depth obtained was

0.0725 m after running for 38 minutes which showed

a good agreement when compared to the experimental

result of 0.0762 m at the same time. This numerical

model can be used for simulations in alignment angle

arrangement.

3 NUMERICAL SIMULATION OF

ALIGNMENT ANGLE

ARRANGEMENTS

The simulations of scour around four cylinders were

numerically performed using Flow-3D software with

inclination angles arrangements of 30°, 45°, and 60°

and various pile spacing ratios, G/D of 2.5, 3, and 3.5

to estimate the depth of sediment scour and assess the

mechanism. All simulation processes are the same as

before in the verification stage. The channel geometry

was modeled as 42 m × 30 m × 7 m (L × W × H) with

an entry length of 100 m before the inlet. The pile

diameter of 1.2 m was located 18 m downstream from

the inlet on a 3 m deep sediment bed with mean

particle diameters d

of 0.002 m, 0.018 m, 0.022 m,

and 0.024 m. The numerical setup at the validation

stage is used for modeling several alignment angle

arrangements with the sizes of minimum and

maximum cells are 0.12 m and 0.60 m. The boundary

conditions were also adopted from the previous setup

with an inlet velocity of 2 m/s. The numerical model

of geometry and meshing for inclination angle

arrangements are shown in Figure 4.

(a)

(b)

Figure 4: Numerical model for angle arrangements with the

display of (a) geometry and (b) meshing.

A total of 9 numerical simulations were carried

out and run for 600 seconds using Flow-3D software,

resulting in the depth of sediment scour as shown

below.

(a)

ICATECH 2023 - International Conference on Advanced Engineering and Technology

142

(b)

(c)

Figure 5: Sediment scour depth with α = 30° and different

G/D (a) 2.5 (b) 3 (c) 3.5.

As seen in Figure 5, the greatest depth of local scour

produced around four piles, respectively from piles 1

to 4 with an incline angle of 30° for G/D = 2.5 are

0.7999 m, 0.7091 m, 0.8010 m, and 0.8940 m. For

G/D = 3 and 3.5, the greatest depths of local scour are

0.8658 m, 0.9046 m, 0.9887 m, 1.0429 m, and 0.9195

m, 0.8319 m, 0.9283 m, 0.9845 m, respectively.

Almost all the results indicate that the maximum

depth of scour increases gradually from the front pile

to the rear pile under various G/D, which is also

marked with the darkest blue color in the scour hole

area. However, when compared between the pile

spacing ratios, the deepest local scour is generated by

G/D = 3. At the same time, the smallest is obtained

by G/D = 2.5 due to the influence of interference

between piles and changes in pile spacing.

(a)

(b)

(c)

Figure 6: Sediment scour depth with α = 45° and different

G/D (a) 2.5 (b) 3 (c) 3.5.

Numerical Simulation of Scouring Around Four Cylindrical Piles with Different Inclination Angles Arrangements

143

The maximum local scour depths obtained around

four piles from piles 1 to 4 with an incline angle of

45° for G/D = 2.5 are 0.8713 m, 0.9799 m, 0.9903 m,

and 1.1682 m. For G/D = 3 and 3.5, the maximum

depths of local scour are 0.8153 m, 1.0544 m, 1.1026

m, 1.1966 m, and 0.8670 m, 0.9300 m, 0.9342 m,

1.0930 m, respectively. Same as the previous results

for α = 30°, because of the accelerated flow and

elevated turbulence, all the results indicate that the

maximum depth of scour increases gradually from the

front pile to the rear pile for all G/D. Compared

between pile spacing ratios, the maximum depth of

local scour is obtained by G/D = 3, and the smallest

is obtained by G/D = 3.5 due to the influence of

interference between piles along with changes in pile

spacing.

(a)

(b)

(c)

Figure 7: Sediment scour depth with α = 60° and different

G/D (a) 2.5 (b) 3 (c) 3.5.

The maximum local scour depths obtained around

four piles from piles 1 to 4 with an incline angle of

60° for G/D = 2.5 are 0.8713 m, 0.9848 m, 1.0995 m,

and 1.2188 m. For G/D = 3 and 3.5, the maximum

depths of local scour are 0.8792 m, 1.0449 m, 1.1017

m, 1.1745 m, and 0.8717 m, 0.9056 m, 0.9720 m,

1.1336 m, respectively. As seen in Figure 7, the

maximum depth of scour increases gradually from the

pile in front to the rear for all G/D. Compared

between pile spacing ratios, the maximum depth of

local scour is obtained by G/D = 2.5. In contrast, the

smallest is obtained by G/D = 3.5, which is slightly

different from G/D = 3 due to the influence of

interference between piles and pile spacing changes.

Compared between angles of attack, the maximum

local scour depth gradually increases with increasing

angle of attack, as stated by the experimental study of

(Zhou et al., 2020). However, some results have

different trends because the maximum scour depths

are not obtained at equilibrium conditions.

4 CONCLUSIONS

The study of sediment scours around four cylindrical

piles with inclination angles arrangements of 30°,

45°, and 60° and pile spacing ratios, G/D of 2.5, 3,

and 3.5 has been performed numerically to estimate

the depth of local scour and investigate its

mechanism. Based on the results, Flow-3D software

is an effective tool for calculating the depth of

sediment scour using the bed load transport rate

equation of Van Rijn and the turbulence model of

RNG k-ε. The greatest depth of local scour obtained

for all inclination angle arrangements at each pile

ICATECH 2023 - International Conference on Advanced Engineering and Technology

144

spacing ratio has the same trend of increasing

gradually from the pile in front to the back due to

accelerated flow and elevated turbulence. However,

when compared between pile spacing ratios, the

smallest maximum scour depth is obtained by G/D =

2.5 for α = 30° and G/D = 3.5 for α = 45° and 60°

because of the interference effect between adjacent

piles. Although most numerical results are linear

experiments, further studies in the alignment angle

arrangements still need to determine and obtain the

maximum scour at the equilibrium depth.

REFERENCES

Brethour, J. (2003). Modeling Sediment Scour - Flow 3D

Technical Notes. 6.

Deng, L., & Cai, C. S. (2010). Bridge Scour: Prediction,

Modeling, Monitoring, and Countermeasures—

Review. PRACTICE PERIODICAL ON STRUCTURAL

DESIGN AND CONSTRUCTION, May, 125–134.

https://doi.org/10.1061/_ASCESC.1943-5576.0000041

Flow Science. (2008). Flow-3D User Manual Version 9.3

(Issue 1). www.flow3d.com

Ghaderi, A., & Abbasi, S. (2019). CFD simulation of local

scouring around airfoil-shaped bridge piers with and

without collar. Sadhana - Academy Proceedings in

Engineering Sciences, 44(10), 216.

https://doi.org/10.1007/s12046-019-1196-8

Hamidi, A., & Siadatmousavi, S. M. (2018). Numerical

simulation of scour and flow field for different

arrangements of two piers using SSIIM model. Ain

Shams Engineering Journal, 9(4), 2415–2426.

https://doi.org/10.1016/j.asej.2017.03.012

Jalal, H. K., & Hassan, W. H. (2020). Three-dimensional

numerical simulation of local scour around circular

bridge pier using Flow-3D software. IOP Conference

Series: Materials Science and Engineering, 745(1).

https://doi.org/10.1088/1757-899X/745/1/012150

Jia, Y., Altinakar, M., & Guney, M. S. (2018). Three-

dimensional numerical simulations of local scouring

around bridge piers. Journal of Hydraulic Research,

56(3), 351–366. https://doi.org/10.1080/00221686.2017

.1356389

Kayser, M., & Gabr, M. (2013). Assessment of scour on

bridge foundations by means of in situ erosion

evaluation probe. Transportation Research Record,

08(2335), 72–78. https://doi.org/10.3141/2335-08

Khosronejad, A., Kang, S., & Sotiropoulos, F. (2012).

Experimental and computational investigation of local

scour around bridge piers. Advances in Water

Resources, 37, 73–85. https://doi.org/10.1016/

j.advwatres.2011.09.013

Kim, H. S., Roh, M., & Nabi, M. (2017). Computational

modeling of flow and scour around two cylinders in

staggered array. Water (Switzerland), 9(9), 1–19.

https://doi.org/10.3390/w9090654

Lu, J.-Y., Hong, J.-H., Su, C.-C., Wang, C.-Y., & Lai, J.-S.

(2008). Field Measurements and Simulation of Bridge

Scour Depth Variations during Floods. Journal of

Hydraulic Engineering, 134(6), 810–821. https://

doi.org/10.1061/(asce)0733-9429(2008)134:6(810)

Nazari-Sharabian, M., Nazari-Sharabian, A., Karakouzian,

M., & Karami, M. (2020). Sacrificial piles as scour

countermeasures in river bridges a numerical study

using FLOW-3D. Civil Engineering Journal (Iran),

6(6), 1091–1103. https://doi.org/10.28991/cej-2020-

03091531

Omara, H., Elsayed, S. M., Abdeelaal, G. M., Abd-

Elhamid, H. F., & Tawfik, A. (2018).

Hydromorphological Numerical Model of the Local

Scour Process Around Bridge Piers. Arabian Journal

for Science and Engineering, 44(5), 4183–4199.

https://doi.org/10.1007/s13369-018-3359-z

Storey, C., & Delatte, N. (2003). Lessons from the Collapse

of the Schoharie Creek Bridge. Forensic Engineering,

158–167. https://doi.org/10.1061/40692(241)18

Tang, J.-H., & Puspasari, A. D. (2021). Numerical

Simulation of Local Scour around Three Cylindrical

Piles in a Tandem Arrangement. In Water (Vol. 13,

Issue 24). https://doi.org/10.3390/w13243623

Van Rijn, L. C. (1984). Sediment transport: bed load

transport. Journal of Hydraulic Engineering - ASCE,

110(10), 1431–1456.

Wei, G., Brethour, J., Grünzner, M., & Burnham, J. (2014).

The Sediment Scour Model in FLOW-3D. FSI-14-

TN9(June), 1–26.

Zhang, Q., Zhou, X. L., & Wang, J. H. (2017). Numerical

investigation of local scour around three adjacent piles

with different arrangements under current. Ocean

Engineering, 142(July), 625–638. https://doi.org/

10.1016/j.oceaneng.2017.07.045

Zhou, K., Duan, J. G., & Bombardelli, F. A. (2020).

Experimental and Theoretical Study of Local Scour

around Three-Pier Group. Journal of Hydraulic

Engineering, 146(10), 04020069. https://doi.org/

10.1061/(asce)hy.1943-7900.0001794

Numerical Simulation of Scouring Around Four Cylindrical Piles with Different Inclination Angles Arrangements

145