International Journal of Fluid Mechanics Research 50(6):1–10 (2023)

CFD INVESTIGATION OF HEAT TRANSFER ON FOUR

CYLINDERS IN IN-LINE SQUARE ARRANGEMENT
EQUIPPED WITH SPLITTER PLATES
Zerrouki Djilali,1 Ladjedel Omar,1 Kahil Yacine,1 Morad Belharizi,2
Adjlout Lahouari,1 Yahiaoui Tayeb,3,* Sikula Ondrej,3 Zemmani Farah,4 &
Belhadj Senini Lina Wafaa4
1
Laboratory of Naval Aero-Hydrodynamics, Mechanical Engineering Faculty, USTO Oran,
Algeria
2
Laboratory of Aeronautics and Propulsion Systems, Mechanical Engineering Faculty, USTO
Oran, Algeria
3
Brno University of Technology, Civil Engineering Faculty, Veveří 331/95 Brno 60200, Czechia
4
Laboratory of Maritime Sciences and Engineering, Mechanical Engineering Faculty, USTO
Oran, Algeria
*Address all correspondence to: Yahiaoui Tayeb, Brno University of Technology, Civil Engineering Faculty,
Veveří 331/95 Brno 60200, Czechia; Tel.: +420 54114 7923, E-mail: [email protected]

Original Manuscript Submitted: 8/10/2022; Final Draft Received: 7/24/2023

A numerical study was carried out for heat transfer in flow past four cylinders in in-line square arrangement with a
pitch-to-diameter ratio L/D = 4.2. Computational fluid dynamics (CFD) computations were performed for Reynolds
numbers of 9300 and 20,000 to investigate passive control using a rectangular splitter plate placed on the rear side
of the cylinders. Three different configurations were considered: the first one contains only smooth cylinders, the
second uses cylinders equipped with splitter plate in the upstream row, and the third configuration where all cylinders
have a splitter plate. The steady-state Reynolds-averaged Navier–Stokes equations were solved using a finite volume
method, where the k-ω SST turbulence model was used to produce a closed system of solvable equations. Results of the
simulations in terms of temperature and Nusselt number distributions for all cylinders are presented and compared.
The effect of new configurations on the heat transfer is well demonstrated, where a drop of about 7.92% in Nusselt
number is reported using the splitter plate.

KEY WORDS: heat transfer, splitter plate, Nusselt number, turbulence

1. INTRODUCTION
Understanding the behavior of fluid flow around cylinder arrays is important in various fields, including aerospace engi-
neering, civil engineering, and renewable energy. The study of fluid flow around a bank of cylinders arranged in different
configurations is a multidisciplinary research area that combines experimental and numerical techniques to gain insights
into complex fluid dynamics and enhance the design and performance of various engineering systems. In industry, the
application often found is the behavior of flow around tube bundles in heat exchangers, particularly when the flow is
transverse. Controlling flow and heat transfer in a tube bundle is important in various industrial applications, such as
heat exchangers and boilers. Several techniques can be employed to optimize flow distribution and enhance heat transfer
within the tube bundle. Convective heat transfer around two tandem circular cylinders has been investigated by research-
ers such as Buyruk (2002), Juncu (2007), Mahir and Altac (2008, 2017), Harimi and Saghafian (2012), in which they

2152-5102/23/$35.00 © 2023 by Begell House, Inc. www.begellhouse.com 1

2 Djilali et al.

conducted in-depth studies to understand the effects of spacing ratio, Reynolds number, and Prandtl number on the fluid
flow pattern as well as the forced convective heat transfer rate. They found that the heat transfer coefficients increase with
the increase of Reynolds number and the distance between the two cylinders. Ladjedel et al. (2017) investigated numeri-
cally heat transfer on two grooved cylinders in tandem arrangement with L/D = 4.2 and Re = 2 × 104. They found that
for the tested Reynolds number, the local Nusselt number for the downstream cylinder exhibits a double hump. A drop
of 15% in the Nusselt number is found when grooves are used. Even though the two-cylinder configurations have been
studied extensively, the four-cylinder configurations have not been widely investigated because of factors like complex-
ity of vortex shedding and the number of geometrical parameters involved. The square configuration of four cylinders is
a commonly studied geometry in the context of flow and heat transfer. Research on the square bundle configuration has
explored various aspects related to flow distribution, pressure drop, and heat transfer characteristics (Lam and Lo, 1992;
Lam and Fang, 1995; Lam et al., 2003a,b). Most of the published experiments have been conducted either in the laminar
regime; Re = 100‒200, mainly for the study of flow visualization, or in the subcritical regime; Re = 103‒104, for pressure,
velocity, drag and lift measurements. On the other hand, numerical simulations have been performed on the four-cylinder
configuration by using different methods and techniques (Lam et al., 2008; Lam and Zou, 2010; Zhao and Cheng, 2012;
Tong et al., 2014; Gao et al., 2017, 2019; Han et al., 2013; Kahil et al., 2018). Several similar experimental and numerical
studies also discussed the flow characteristics for four cylinders, square (Liu et al., 2015; Zhang et al., 2021), elliptical
(Puliyeri et al., 2021), and grooved (Ladjedel et al., 2015). Their results confirm that the flow characteristics of multiple
cylinders strongly depend on geometrical parameters and the attack angle, which affect the stability and the complexity
of flow fields and enhance the interactions between cylinders. Some numerical studies have been performed, for example,
by Esfahani and Vasel-Be-Hagh (2012), using the lattice Boltzmann method to determine heat transfer characteristics at
various Reynolds numbers. They found that the maximum heat transfer rate of the square unit is related to the stagnation
point of the upstream cylinder, where the mean Nusselt number is higher. They also found that if the Reynolds number
increases, the thermal shear layer thickness decreases for all cylinders. Kim (2013) investigated the effect of the longitu-
dinal pitch on the single-phase heat transfer characteristics in cross flow over in-line tube banks. He found that an aspect
ratio greater than 2.7 does not affect heat transfer, although the reduction of pitch ratios less than 2.7 leads to a reduction
in heat transfer.
The flow characteristics of multiple cylinders in a configuration, such as the presence of a splitter plate, are influ-
enced by various geometrical parameters. These parameters play a crucial role in determining the flow patterns, pressure
drop, and heat transfer performance within the cylinder group. Many researchers have investigated the effect of splitter
plate on fluid characteristics such as the vortex shedding, pressure distributions, and drag and lift coefficients (Akilli et
al., 2005, 2008; Vu et al., 2015; and An et al., 2019). Razavi et al. (2008) investigated the effect of a splitter plate length
on flow-induced forces and the heat transfer behavior of a circular cylinder at low Reynolds numbers (20 < Re < 1000).
A significant reduction in the drag coefficient as well as the average Nusselt number was observed in the presence of a
splitter plate, implying stabilization of the wake region and reduction of the vortex shedding. However, the conductive
heat transfer was increased as a result of the extra heat transfer area generated by the splitter plate, upon which the overall
heat transfer of the system was improved.
The purpose of this study is to investigate numerically the heat transfer mechanism and to exhibit the influence of
controlling the flow over square in-line four cylinders. The square bundle configuration serves as a benchmark case for
flow and heat transfer studies, allowing researchers to investigate fundamental phenomena and develop practical design
guidelines. The use of a splitter plate in a group of cylinders could indeed have a significant effect on heat transfer en-
hancement. It could alter the flow patterns, promote better mixing, and enhance heat transfer within the cylinder group.
The findings from these studies can be extrapolated and applied to other tube bundle configurations or more complex
geometries.

2. NUMERICAL PROCEDURE
In the present study, the dimensionless governing equations with the appropriate boundary conditions are solved us-
ing the finite volume method (FVM) by employing a structured grid. The commercial CFD software Ansys−Fluent is
used to solve the governing equations. The semi-implicit method for pressure-linked equations (SIMPLE) algorithm
is used to determine the pressure field, while the quadratic upwind interpolation for convective kinematics (QUICK)

International Journal of Fluid Mechanics Research

Heat Transfer on Four Cylinders in In-Line Square Arrangement Equipped with Splitter 3

scheme is used to discretize the convection terms in the momentum and energy equations. The Reynolds-averaged
Navier–Stokes RANS equations for continuity and momentum conservation can be written as

U j
0 (1)
x j


   U i  P
(U jU i )    u j u j   (2)
x j x j  x j  xi


The local Nusselt number is evaluated since the conductive heat transfer rate of the fluid over solid boundary
equals its convective heat transfer rate, which can be written as follows:

 T 
k    h (Tw  T )  (3)
 n  wall

h D
Nu   (4)
k

The heat transfer around cylinders is governed by the following expression:

C p (V *T * )  k  2T *  0  (5)

The average Nusselt number is obtained by integrating this value over the surface of the cylinder, which is defined
as


1
Nu  Nu ()d   (6)
 0

The k-ω SST turbulence model is used, which can be written as:

k
t  *  (7)


Dk U i     k
  ij  * f* k    t  x  (8)
Dt x j x j  k  j

Volume 50, Issue 6, 2023

4 Djilali et al.

D  U t     
   ij  fk 2   t   (9)
Dt k x j x j    x j

where the specific dissipation rate is

 1
  . (10)
k 

The present study is limited for subcritical Re of 9.3 × 103 and 2 × 104, as it is also adopted by other authors for
similar studied problems with identical parameters. Reynolds numbers (Re = DU∞/ν) have been based on the oncom-
ing free-stream velocities U∞ of 3.25 and 7.5 m/s, respectively, while the outlet boundary is considered as pressure
outlet having atmospheric pressure along the boundary. The flowing fluid is air with an imposed inlet temperature
Tinlet = 298.15 K and constant physical properties at Prandtl number Pr = 0.71. As shown in Fig. 1, the four circular
cylinders are placed in an in-line square arrangement and maintained with a constant surface temperature Twall = 318.15
K. Longitudinal and transverse distances between the center points of two adjacent cylinders are 4.2D, where D = 40
mm. Figure 2 illustrates the shape and geometrical dimensions of the splitter plate, and Table 1 summarizes the differ-
ent configurations tested in the present study. To be consistent with the literature and to ensure that the occurrence of
vortex shedding can be observed for a range of 9.3 × 103 < Re < 2 × 104, the distance between the center points of two
adjacent circular cylinders is 4.2D, which enables comparison with the literature (Ladjedel et al., 2015, 2017). Since
the effect of the arrangement of four square cylinders on the flow and heat transfer is mainly highlighted in this work,
when the thermal buoyancy is imposed, the selection of the above parameters is reasonable to explore the interactions
between the four square cylinders. Splitter plate length was taken to be L = 5 mm = 0.125D, and its thickness was set S
= 3 mm = 0.075D. This choice is justified in that Gerrard (1966) showed that vortex shedding frequency decreased at
Re = 2 × 104 when the splitter length L is shorter than the cylinder’s diameter, L/D < 1, but it increased for 1 < L/D < 2.

FIG. 1: Computational domain with boundary conditions

International Journal of Fluid Mechanics Research

Heat Transfer on Four Cylinders in In-Line Square Arrangement Equipped with Splitter 5

FIG. 2: Computational grid and closer view of the splitter plate geometry

TABLE 1: Tested cases

Cylinder 1 Cylinder 2 Cylinder 3 Cylinder 4
Case 0 Smooth Smooth Smooth Smooth
Case 1 With splitter plate Smooth With splitter plate Smooth
Case 2 With splitter plate With splitter plate With splitter plate With splitter plate

The inlet boundary conditions are placed at 20 × D for the cylinders. A preliminary calculation was carried out for
a long tube to determine the inlet velocity profile. The latter was taken as the inlet boundary condition. Several grids
have been tested to investigate the influence of grid refinement on the solution; hence, three meshes are employed
Table 2. The computational domain is divided into a number of unstructured hexahedral grids. The grid is nonuniform
in the x–y plane (Fig. 2). There is a very fine spacing near the tube walls, as needed for accurately resolving the steep
gradients in the thin boundary layer. The grid expansion factor following the y direction is set to 1.1 and was used to
investigate the effect of refinement of the grid in the boundary-layer region formed by eight layers.

3. RESULTS AND DISCUSSION

In this section, obtained results for two-dimensional simulations conducted at Reynolds numbers of 9.3 × 103 and 2 ×
104 are presented. For the specified rectangular splitter plate, the variation of local and average Nusselt numbers are
investigated and analyzed. Results are compared to the case with only smooth cylinders, hereafter referred to as case
0. A mesh independence test was performed in order to study the effect of grid size on the results. Three grids (coarse,
medium, and refined) were tested for case 0 with isothermal smooth cylinders in in-line square arrangement at Re = 2
× 104, and the local Nusselt number distributions for cylinder 1 are shown in Fig. 3. According to the above literature,
cylinders at the upstream row behave as isolated cylinders depending on their spacing. Hence it is clearly evident that
cylinder 1 has the same trend as an isolated cylinder in the present configuration. Despite the difference between grids,
results obtained with the medium one show good agreement with results obtained by Buyruk (2002) and Ladjedel et
al. (2015). Therefore, this grid will be used in all subsequent computations.

TABLE 2: The meshes used

Coarse Medium Fine
Node number 201350 400600 806400

Volume 50, Issue 6, 2023

6 Djilali et al.

FIG. 3: Mesh independence test: (a) mesh independence test and validations for an isolated cylinder (Re = 9300) and (b) mesh
independence test for cylinder 1 (Re = 20,000).

Figures 4 and 5 show the variation of local Nusselt number along a half cylinder at Re of 9.3 × 103 and 2 × 104,
respectively. At Re of 9.3 × 103, local Nusselt number reaches its maximum value at the stagnation point θ = 0 for
both cases 1 and 2. Then it slowly decreases until θ = 80°–100°, as it sharply increases reaching a peak value. This
phenomenon occurs in the splitter plate location, which influences heat transfer characteristics at the rear side of
cylinders. At Re 2 × 104, the stagnation point shifts towards the rear side of cylinders for both cases 1 and 2, with
the same trend of local Nusselt number distributions. It is also observed that as the Reynolds number increases, the
separation point shifts towards the rear side of all cylinders. It can be noted that local and average Nusselt numbers
variations are proportional to the Reynolds number variation. Tables 2 and 3 confirm the above observations, and
it can be seen that the average Nusselt number for the first row is lower than that of the second row. This could be
explained by the fact that the boundary layer around the second row seems thinner because of the high turbulence
in the wake region.
Figures 6 and 7 present the temperature contours for case 2 (all cylinders with splitter plate) compared to case 0
(all smooth cylinders). It is clearly observed that the flow topology is influenced by adding splitter plates, where the

FIG. 4: Local Nusselt number distributions for Re = 9300

International Journal of Fluid Mechanics Research

Heat Transfer on Four Cylinders in In-Line Square Arrangement Equipped with Splitter 7

FIG. 5: Local Nusselt number distributions for Re = 20,000

TABLE 3: Average Nusselt number of the studied cases for Re = 9300

Cylinder 1 Cylinder 2 Cylinder 3 Cylinder 4
Case 0 36.064638 31.226445 32.76702 38.094735
Case 1 35.079265 28.840569 32.209804 31.692479
Case 2 35.616632 28.044017 28.361129 35.569475

FIG. 6: Temperature contours for Re = 20,000

flow structures are changed. The initial temperature of the cylinder wall is 398.15 K, which decreases from the stagna-
tion point to the separation point for the whole front side of the cylinder. This temperature gradient is an evidence of
heat transfer between the fluid and the front sides of cylinders. At the rear sides of cylinders with splitter plate, a steep
temperature gradient is observed due to the presence of recirculating vortices in the wake region, which enhance heat
transfer between cylinders and the flowing fluid.
Table 4 shows the total Nusselt number comparison between the studied cases for the investigated Reynolds num-
bers. The table also shows the Nusselt number reduction rate compared to case 0 without passive control on cylinders.
It can be observed that the heat transfer is lower for case 2 compared to case 1. It is also seen that the reduction rate
increases with the increase of Reynolds number, as observed by Ladjedel et al. (2015).

Volume 50, Issue 6, 2023

8 Djilali et al.

FIG. 7: Temperature contours for Re = 9300

TABLE 4: Average Nusselt number of the studied cases for Re = 20,000

Cylinder 1 Cylinder 2 Cylinder 3 Cylinder 4
Case 0 118.1179 117.93522 109.42448 123.01553
Case 1 109.78392 92.10633 91.639319 114.13236
Case 2 104.21102 104.28532 104.48136 109.7224

In general, the presence of splitter plates on the two front cylinders in a four-cylinder configuration does
not significantly influence the Nusselt number results for subcritical Re. The Nusselt number is a dimensionless
parameter used to determine the convective heat transfer coefficient, and it depends on various factors such as
the fluid flow conditions, geometry, and surface characteristics. For subcritical Re, which represents relatively
low flow velocities, the flow around the cylinders is predominantly laminar. In this flow regime, the presence of
splitter plates on the front cylinders might slightly alter the flow patterns and pressure distribution, but the overall
impact on the Nusselt number is generally minimal. Splitter plates are often used to control the flow separation
and reduce the drag on bluff bodies such as cylinders. They can help in enhancing the heat transfer performance,
especially at higher Re, where the flow is turbulent. However, for subcritical Re, the impact of splitter plates on
the Nusselt number is usually insignificant. It’s important to note that the specific geometry, size, and arrangement
of the cylinders, as well as other factors such as spacing, orientation, and Re range, may introduce some variations
in the results (Table 5). Therefore, it’s always advisable to conduct detailed experimental or computational studies
to accurately assess the influence of splitter plates on the Nusselt number in specific configuration and operating
conditions.

TABLE 5: Total Nusselt number with the reduction rate

Total Nu Reduction Rate of Nu
Re = 9300 Case 0 138.152838 —
Case 1 127.822117 7.47774%
Case 2 127.591253 7.64485%
Re = 20,000 Case 0 468.49313 —
Case 1 422.7001 5.87%
Case 2 407.618899 7.92%

International Journal of Fluid Mechanics Research

Heat Transfer on Four Cylinders in In-Line Square Arrangement Equipped with Splitter 9

4. CONCLUSION
A passive control with heat transfer was investigated along four cylinders in square arrangement. Two cases were
tested for Reynolds numbers of 9300 and 20,000. The k-ω SST model was used for simulation. The study quantified
the effect of using the splitter plate as a passive control on heat transfer characteristics in terms of Nusselt numbers as
well as temperature fields. According to the results, the study observed a reduction of 7.92% in the total Nusselt num-
ber when a splitter plate was used at a Reynolds number of 20,000. The total Nusselt number quantifies the convective
heat transfer from a surface and is typically calculated by integrating the local Nusselt number over the entire surface.
The reduction of 7.92% indicates that the presence of the splitter plate resulted in a decrease in the overall convective
heat transfer rate compared to the case without the splitter plate. This reduction in the Nusselt number suggests that
the implementation of the splitter plate negatively affected the convective heat transfer characteristics. As mentioned
earlier, the splitter plate may have suppressed vortex shedding and flow-induced forces, resulting in a decrease in
the heat transfer coefficient and subsequently the total Nusselt number. It is important to consider that these results
are specific to the conditions and setup of the study. The observed drop in the Nusselt number may vary for different
Reynolds numbers, geometries, and operating conditions. Further analysis and experimentation would be required to
understand the underlying mechanisms and optimize the design of the splitter plate for heat transfer enhancement in
the specific system.

ACKNOWLEDGMENT
This work was supported by specific research at the Brno University of Technology through Project No. FAST-
J-19-5976, “Optimization of thermally activated structures.”

REFERENCES
Akilli, H., Sahin, B., and Tumen, N.F., Suppression of Vortex Shedding of Circular Cylinder in Shallow Water by a Splitter Plate,
Flow Meas. Instrum., vol. 16, no. 4, pp. 211–219, 2005.
Akilli, H., Karakus, C., Akar, A., Sahin, B., and Tumen, N.F., Control of Vortex Shedding of Circular Cylinder in Shallow Water
Flow Using an Attached Splitter Plate, J. Fluids Eng., vol. 130, no. 4, p. 41401, 2008.
An, X., Song, B., Tian, W., and Ma, C., Numerical Research of Flow Past a Circular Cylinder with Splitter Plate at a Subcritical
Reynolds Number Region, J. Shanghai Jiaotong Univ. (Sci.)., vol. 24, pp. 113–121, 2019.
Buyruk, E.R.T.A.N., Numerical Study of Heat Transfer Characteristics on Tandem Cylinders, Inline and Staggered Tube Banks in
Cross-Flow of Air, Int. Commun. Heat Mass Transf., vol. 29, no. 3, 2002.
Esfahani, J.A. and Vasel-Be-Hagh, A., LB Simulation of Heat Transfer in Flow Past a Square Unit of Four Isothermal Cylinders,
C. R. Mec., vol. 340, no. 7, pp. 526–535, 2012.
Gao, Y.Y., Yin, C.S., Zhang, H.Q., Yang, K., Zhao, X.Z., and Sun, Z., Numerical Study on Flow around Four Square-Arranged
Cylinders at Low Reynolds Numbers, Math. Probl. Eng., Article ID 6381256, 2017.
Gao, Y., Yang, K., Zhang, B., Cheng, K., and Chen, X., Numerical Investigation on Vortex-Induced Vibrations of Four Circular
Cylinders in a Square Configuration, Ocean Eng., vol. 175, pp. 223–240, 2019.
Gerrard, J.H., The Mechanics of the Formation Region of Vortices Behind Bluff Bodies, J. Fluid Mech., vol. 25, no. 2, pp. 401–413,
1966.
Han, Z., Zhou, D., Gui, X., and Tu, J., Numerical Study of Flow Past Four Square-Arranged Cylinders Using Spectral Element
Method, Comput. Fluids, vol. 84, pp. 100–112, 2013.
Harimi, I. and Saghafian, M., Numerical Simulation of Fluid Flow and Forced Convection Heat Transfer from Tandem Circular
Cylinders Using Overset Grid Method, J. Fluids Struct., vol. 28, pp. 309–327, 2012.
Juncu, G., A Numerical Study of Momentum and Forced Convection Heat Transfer around Two Tandem Circular Cylinders
at Low Reynolds Numbers, Part I: Momentum Transfer, Int. J. Heat Mass Transf., vol. 50, nos. 19–20, pp. 3788–3798,
2007.
Kahil, Y., Benhamadouche, S., Berrouk, A.S., and Afgan, I., Simulation of Subcritical-Reynolds-Number Flow around Four Cylin-
ders in Square Arrangement Configuration Using LES, Eur. J. Mech. B Fluids, vol. 74, pp. 111–122, 2018.
Kim, T., Effect of Longitudinal Pitch on Convective Heat Transfer in Crossflow over In-Line Tube Banks, Ann. Nuclear Energy,
vol. 57, pp. 209–215, 2013.

Volume 50, Issue 6, 2023

10 Djilali et al.

Ladjedel, O., Adjlout, L., Yahiaoui, T., Imine, O., and Sikula, O., Numerical Investigation of Heat Transfer on Two Grooved Cyl-
inders in a Tandem Arrangement, Comput. Therm. Sci.: Int. J., vol. 9, no. 6, 2017.
Ladjedel, O., Bouacha, M., Adjlout, L., Yahiaoui, T., and Belharat, N., Experimental Investigation of Turbulent Flow Past Four
Grooved and Smooth Cylinders in an In-Line Square Arrangement, in EPJ Web of Conferences (vol. 92, p. 02046), EDP Sci-
ences, 2015.
Lam, K. and Lo, S.C., A Visualization Study of Cross-Flow around Four Cylinders in a Square Configuration, J. Fluids Struct., vol.
6, no. 1, pp. 109–131, 1992.
Lam, K. and Fang, X., The Effect of Interference of Four Equispaced Cylinders in Cross Flow on Pressure and Force Coefficients,
J. Fluids Struct., vol. 9, no. 2, pp. 195–214, 1995.
Lam, K., Li, J.Y., Chan, K.T., and So, R.M.C., Flow Pattern and Velocity Field Distribution of Cross-Flow around Four Cylinders
in a Square Configuration at a Low Reynolds Number, J. Fluids Struct., vol. 17, no. 5, pp. 665–679, 2003a.
Lam, K., Li, J.Y., and So, R.M.C., Force Coefficients and Strouhal Numbers of Four Cylinders in Cross Flow, J. Fluids Struct., vol.
18, nos. 3–4, pp. 305–324, 2003b.
Lam, K., Gong, W.Q., and So, R.M.C., Numerical Simulation of Cross-Flow around Four Cylinders in an In-Line Square Configu-
ration, J. Fluids Struct., vol. 24, no. 1, pp. 34–57, 2008.
Lam, K. and Zou, L., Three-Dimensional Numerical Simulations of Cross-Flow around Four Cylinders in an In-Line Square Con-
figuration, J. Fluids Struct., vol. 26, no. 3, pp. 482–502, 2010.
Liu, M., Xiao, L., and Yang, L., Experimental Investigation of Flow Characteristics around Four Square-Cylinder Arrays at Sub-
critical Reynolds Numbers, Int. J. Nav. Archit. Ocean Eng., vol. 7, no. 5, pp. 906–919, 2015.
Mahir, N. and Altaç, Z., Numerical Investigation of Convective Heat Transfer in Unsteady Flow Past Two Cylinders in Tandem
Arrangements, Int. J. Heat Fluid Flow, vol. 29, no. 5, pp. 1309–1318, 2008.
Mahir, N. and Altaç, Z., Numerical Investigation of Flow and Heat Transfer Characteristics of Two Tandem Circular Cylinders of
Different Diameters, Heat Transf. Eng., vol. 38, no. 16, pp. 1367–1381, 2017.
Puliyeri, V., Vengadesan, S., and Arul Prakash, K., Effect of Cylinder Arrangement on Fluid Flow and Heat Transfer Characteristics
Past Four Elliptic Cylinders, Heat Transf. Eng., vol. 42, no. 21, pp. 1789–1810, 2021.
Razavi, S.E., Farhangmehr, V., and Barar, F., Impact of a Splitter Plate on Flow and Heat Transfer around Circular Cylinder at Low
Reynolds Numbers, J. Appl. Sci., vol. 8, no. 7, pp. 1286–1292, 2008.
Tong, F., Cheng, L., Zhao, M., Zhou, T., and Chen, X.B., The Vortex Shedding around Four Circular Cylinders in an In-Line Square
Configuration, Phys. Fluids, vol. 26, no. 2, 2014.
Vu, H.C., Ahn, J., and Hwang, J.H., Numerical Investigation of Flow around Circular Cylinder with Splitter Plate, KSCE J. Civ.
Eng., vol. 20, pp. 2559–2568, 2015.
Zhang, Y., Xie, G., Sundén, B., and Li, Y., An LBM-Based Investigation of Thermal Buoyancy and Arrangement Angle on Flow
Characteristics and Heat Transfer over Four Heated Square Cylinders, Numer. Heat Transf., Part B, vol. 79, nos. 5–6, pp.
278–301, 2021.
Zhao, M. and Cheng, L., Numerical Simulation of Vortex-Induced Vibration of Four Circular Cylinders in a Square Configuration,
J. Fluids Struct., vol. 31, pp. 125–140, 2012.

International Journal of Fluid Mechanics Research