Pertanika J. Sci. & Technol.

28 (1): 1 - 18 (2020)

SCIENCE & TECHNOLOGY

Journal homepage: http://www.pertanika.upm.edu.my/

Analysis of Velocity Profiles in Rectangular Straight Open

Channel Flow

Abinash Sahoo1*, Sandeep Samantaray1 and Rosysmita Bikram Singh2

Department of Civil Engineering, National Institute of Technology Silchar, Assam 788010, India
1

Department of Civil Engineering, Veer Surendra Sai University of Technology Burla, Odisha 768018, India
2

ABSTRACT
This study aims at analysing the velocity profiles in a straight rectangular channel having
a constant width and depth throughout the channel length. Experiment was conducted
for five altered depths using the straight rectangular flume in hydraulic flow laboratory
available at VSS University of Technology, Odisha, India. Furthermore, the distributions of
stream wise velocity at the different flow depth were computed numerically using hydraulic
software ANSYS FLUENT 18.1. The results of numerical simulation showed sensibly
good agreement with experimental data, which was ±10% for both across and along the
channel of rectangular flume. The Reynolds number in this study lay between10,853 to
79,000. In case of channel flow, the velocity varied both in longitudinal and transverse
direction. The isovel lines joining points of equal velocity normally curved upward due to
the effect of turbulence. Peak velocity was found below the free surface of water. The law of
logarithmic and power law was applied to study velocity distribution in terms of turbulent
flow condition. The model could be validated by considering the various parameters of
flow measurement such as the resultant velocity and the velocity at both horizontal and
vertical direction. The inclusive idea of this study was to comprehend flow characteristic
over a plane bed through experimental and numerical simulation methods.
Keywords: ANSYS, logarithmic law, power law, rectangular flume, velocity distribution, velocity profile

INTRODUCTION

ARTICLE INFO
The distribution of longitudinal component
Article history: of velocity in a given cross-section is a
Received: 13 January 2019
Accepted: 19 July 2019 significant parameter in the study of an open
Published: 13 January 2020
channel flow to find out various properties
E-mail addresses:
[email protected] (Abinash Sahoo) of flow. Diverse and complex behaviour
[email protected] (Sandeep Samantaray)
[email protected] (Rosysmita Bikram Singh)
of fluid flow which occurs at a particular
* Corresponding author period generates intricate structures that

ISSN: 0128-7680
e-ISSN: 2231-8526 © Universiti Putra Malaysia Press
 Abinash Sahoo, Sandeep Samantaray and Rosysmita Bikram Singh

have to be modelled or solved partially based on time and length scale. Hence a study of
a consistent mathematical model of velocity distribution is frequently needed. Study of
velocity distribution pattern indicates there is a rise in velocity consistently with vertical
distance from bed of the channel. At lower depths of flow, magnitude of velocity is least
at the bed because of no slip condition while velocity is found to be highest at some point
below the free surface. When there is an increase in depth, the sequences of greater velocity
occurs at free surface of the wall.
Bonakdari et al. (2014) developed a 3D Computational Fluid Dynamics (CFD) model
using ANSYS software to examine flow patterns and the k-ε turbulence closure model
was applied to resolve turbulence equations. Wake function is unable to detect the action
of secondary currents and thus is not applicable to 3D flows, as it helps in predicting an
increasing velocity with distance from the bed. Coles (1956) found that k-ε model took into
account two equations for isotropic turbulence. Hence, model is inadequate to speculate
the secondary currents and velocity-dip-phenomenon related to it. Nezu and Rodi (1986)
and Cardoso et al. (1989) applied the classical log law to determine velocity profile for
vertical flow in an open channel at a section ξ<0.2{ξ = y/h, is ratio between distance from
bed(y) to flow depth (h)}. Castro-Orgaz, and Dey (2011) proposed theory of power-law for
velocity profile for flow under turbulent boundary layer. Naik et al. (2018) used ANSYS-
Fluent software to simulate the model in two phases numerically in a converging compound
channel. Welahettige et al. (2017) used ANSYS Fluent R16.2 for CFD simulations and
outcomes were validated with experimental results done in an open Venturichannel. Naot
et al. (1993) applied Log-wake and power law to narrate flows in two-dimensional open-
channel. Kang and Choi (2006) gave additional advanced RANS equations based on
anisotropic turbulence models and should be used for prediction of velocity-dip-phenomena.
Sarma et al. (2000) described velocity distribution taking into consideration velocity dip
for flow in open channel. They used binary version of velocity distribution, where inner
section follows logarithmic law and outer section follows parabolic law. Afzal et al. (2007)
used Velocity profile power law to cover the influence of friction in fully developing zone
for turbulent pipe and open channel flows. Wilkerson and McGahan (2005) expanded two
models which help in predicting depth-averaged velocity distributions. Guo and Julien
(2006) derived modified-log-wake law to predict velocity-dip-phenomenon for flow in
open channel. Absi (2011) modified this law and proposed another law for velocity profile
called the full dip-modified-log-wake law and law was semi analytical.
The purpose of this study is to compute magnitude of velocity at various points
longitudinally and laterally of the rectangular flume, which assists to draw the velocity
flow profiles that help in studying the velocity distribution and its characteristics. The
resulting velocity profiles are validated with the universal laws. The results found from
this computational approach are hence contrasted with the experimental data for better

2 Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020)

 Analysis of Velocity Profiles in Rectangular Straight Open Channel Flow

efficiency. The computational way is considerably being used globally due to competency
and flow complexity, which helps in resolving and modelling instinctively in accordance
to what the model requires.

METHODS
Theoretical Consideration
Study from literature suggested that in an open channel the magnitude of velocity is
maximum at some point below top surface of water. For narrow channels where the aspect
ratio Ar<5, the occurrence of maximum velocity is below free surface of water. By assuming
“shear stress is constant”, P-vK (Prandtl-von Karman) law is considered as general form of
velocity distribution, applied near channel bed. The von Karman constant is modified and
it can be applied in outer region of flow. P-vK law overlaps with log law at about 20% of
flow depth and can be used for full flow depth. Log law is strictly applicable to the region
below 20% of flow depth that may be attributed to the law of the wall (log-law) which is
often applied to the entire depth. Log law is effective for inner region near the bed, but it
will be deviated from the laboratory data on the outer layer:

u 1 y
= [Ln � (1)
u∗ k y0

Where, u = mean velocity in main flow direction

u* =shear velocity
y = upright distance from bed
y0= distance where velocity is hypothetically zero
K denotes the von-Karman constant. For outer region of flow i.e. ξ > 0.2, there is a
variance of log law from experimental data. So, for two-dimensional flows, this variance
was adjusted by summing Coles wake function as

u 1 y πy
= [ln + 2Πsin2 ] (2)
u∗ k y0 2h

Where h = flow depth

𝛱 = Coles’ wake strength. The expression is called the log wake law, notation П express
the Coles parameter, denoting the wake function strength. General form of parabolic law
given by:
u−umax η
= 6.3(1 − )2 (3)
u∗ ηdip

Where η· = dimensionless distance from bed

η d i p = dimensionless distance of maximum velocity.

Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020) 3

 Abinash Sahoo, Sandeep Samantaray and Rosysmita Bikram Singh

On basis of the intersection point, the parabolic law is valid to the free surface which
ranges from 0.2 to 0.3 for flows in sediment-laden. Later, Coles proposed the limitations
of the law and validity region for parabolic law. For a narrow channel where the aspect
ratio Ar<5, peak velocity occurs below free surface of water, resulting in phenomenon of
velocity-dip-, which involves a variance from log-wake law. Using the laws proposed by
Coles, velocity profile was validated in this present study.

Experimental Setup
Experiments were conducted utilizing the flume present in the Hydraulics Lab of VSSUT,
Burla as shown in Figure 1. The experiments were performed in a straight simple rectangular
flume described below in Figure 2 made of iron whose length is 2.5 m, width 0.3 m, and
0.3 m deep. The water supply was from a storage sump which was lifted by a number of
parallel pumps to an overhead tank. From the overhead tank water was carried to the flume
and discharge was controlled through a valve. The head gate was lifted fully to allow the
water to enter the flume. Baffle walls were provided in series before the head gate so as to
decrease the turbulence effect of the incoming water. The Point gauge was first checked
for its smooth moving over the flume and its least count was noted. The tail-gate was
responsible to maintain the uniformity of flow which was accumulated in a metal tank,

Open sump Volumetric Tank

WATER
CIRCULATION

Motor Pump Straight Flume

Overhead Tank
Figure 1. Experimental setup in the Lab

4 Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020)

 Analysis of Velocity Profiles in Rectangular Straight Open Channel Flow

where the volumetric measurement could be taken. Further it can be redirected to the sump
and from sump to overhead tank, with a help of a pump installed in the laboratory, thus
setting a complete re-circulating water supply system to the flume. To attain a steady and
uniform flow conditions every single experimental run of the flume was done by keeping
the water surface gradient parallel to the bed slope.
The flow velocity in the experiment were measured at 9 mid-verticals (x = 0.25, 0.5,
0.75, 1.0, 1.25, 1.5, 1.75, 2.0, 2.25 m) alongside the flow developing region and 6 mid-
horizontals (y = 0.03, 0.06, 0.09, 0.12, 0.15 and 0.18 m) across the flow developing zone
across half width of the flow section revealed in Figure 2. This was for a single water flow
depth. In this way, the same procedure was repeated for multiple flow depths starting from
bed of the channel (0.06, 0.1, 0.14, 0.18, 0.22 m).

Open Sump
Overhead Tank

Volumetric Tank

Head Gate Travelling Bridge Tail Gate

Baffle walls
(perforated walls) FLUME

Figure 2. Plan view of the experimental flume

Table 1
Details of the experimental conditions

Test Flow Discharge, Q Height, h Aspect Ratio, Froude number Reynolds number
No. (cm3/s) (m) b/h (Fr) (Re)
1 19.55 0.06 5 0.29 10,853
2 27.2 0.1 3 0.323 21,573
3 38.25 0.14 2.14 0.384 36,404
4 51 0.18 1.66 0.451 54,606
5 67.15 0.22 1.36 0.537 79,000

Table 1 indicates the different heights from channel bed where the velocity
measurements have been conducted with different aspect ratio. Fr -Froude number = 𝑉 ⁄ 𝑔ℎ
, where V-average velocity, Re -Reynolds number = VR⁄ν where R-hydraulic radius.

Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020) 5

 Abinash Sahoo, Sandeep Samantaray and Rosysmita Bikram Singh

Depth of flow in rectangular channel was measured through piezometer which was
static on the traveling bridge and manually functioned. We can note the height of water
entering into the tank and with the help of a stopwatch we can calculate the time. Dividing
height by time we obtained the velocity and then multiplying it with the base area of the
tank we obtained the average discharge. The velocity of the flow was measured at various
points using a Pitot tube connected to a manometer. While noting velocity readings using
Pitot tube, Pitot was positioned facing towards the flow direction. After this setup it was
again turned across flow direction. Along the longitudinal section of rectangular flume pitot
was moved at 0.25 m intervals in horizontal direction and 0.2h, 0.4h, 0.6h, 0.8h (h-flow
depth) just below free surface of the respective flow depth in vertical direction. Similarly,
along lateral section of the flume pitot was moved at 0.03m intervals in horizontal direction
and 0.2h, 0.4h, 0.6h, 0.8h below free surface in vertical direction as depicted in Figure
3. As the cross-section of channel was square in shape, velocity was measured from near
side wall till mid interval of the channel because the magnitude of velocity remained same
on the other half of the interval. The Pitot tube was firmed to primary scale which had
least count of 0.1 mm with the Vernier scale. To the right limb of the Pitot the pipe was
attached to the right limb of the manometer and the left limb of the pitot to the left limb.
The velocity hence was calculated by formula:

𝑉 = 2𝑔∆𝐻 (4)
where g -force of gravity, ΔH -water elevation difference in the manometer.

0.30m Near boundary

b/2 = 0.15m

b = 0.30m 0.03m
0.15m At center line
0.25m
Left = 2.5m

Near top

0.8h

06h

0.4h

0.2h

0.03m
b/2 = 0.15m

Figure 3. Detail plan (a) and cross-section (b) of experimental grid where velocity measured

6 Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020)

 Analysis of Velocity Profiles in Rectangular Straight Open Channel Flow

NUMERICAL MODEL SETUP

A flow through the open channel is dependent on the various features such as bed properties,
shear resistance and friction. k-ε model is highly adopted model in CFD to simulate
characteristics of flow in turbulent flow condition. Larocque et al. (2013) proposed k-ε
model which emphasised on mechanisms that impacted on turbulent kinetic energy. This
kind of generality is lacking in the mixing length model. Assumptions made for the model
is ‘the average rate of distortions is alike in all directions’. It is a semi-empirical model
which depends on model transport equations for turbulent-kinetic energy ‘k’ and dissipation
rate ‘ε’. Simulation of flow model can be solved in an efficient manner with the help of
ANSYS (Fluent). The cross-section of the rectangular flume used for the experimental
purpose is drawn with the help of ANSYS software and is divided into different sections
with plates placed both in horizontal and vertical directions as revealed in Figure 4 for
numerical analysis. The fluid flow equation is governed by momentum equation and
continuity equation. This method is fit for structured as well as unstructured mesh.

Figure 4. Cross-sectional plates along and across the flow direction

Boundary Conditions
Near wall and channel bed the components of velocity become zero due to effect of no slip
condition. For the free surface, the boundary conditions are symmetric that is no scalar flux
takes place across the boundary. The truncation error that arises by stepwise approximation
can be controlled by providing a very fine Cartesian mesh. Nodes are required for the area
near the wall and the wake regions. For this case study under consideration the flow domain
was discretized by the use of structured grid and body-fitted coordinates. The meshing
details of flume are indicated in Figure 5. Least size of the grid after meshing is 0.2527
mm and maximum size of the grid is found be 50.55 mm. The maximum face size of the
flume is kept 25.27 mm. Grid size may vary depending on different model setup.

Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020) 7

 Abinash Sahoo, Sandeep Samantaray and Rosysmita Bikram Singh

Figure 5. Meshing of the Flume in 3-D ANSYS

Application of Numerical Analysis

In present study, the coupling among velocity field and pressure field is complemented
by PISO technique, which is pressure implicit splitting operator available in Fluent as
proposed by (Issa, 1986). Figure 6, 7 and 8 show the velocity distribution in stream-wise,
vertical, lateral direction and the resultant velocity distribution at 0.25, 1.25 and 2.25 m
from the head gate for no load conditions as simulated in ANSYS. In the present work, it is
observed that the results obtained from ANSYS were matching well with the experimental
data and magnitude of maximum velocity and velocity distribution were virtually equal to
the results derived experimentally.

u v

w R

Figure 6. Contours for stream wise velocity (u), vertical velocity (v), lateral velocity (w), resultant velocity
(R) at 0.25m from head

8 Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020)

 Analysis of Velocity Profiles in Rectangular Straight Open Channel Flow

u v

w R

Figure 7. Contours for stream wise (u), velocity vertical velocity (v), lateral velocity (w), resultant velocity
(R) at 1.5m from head

u v

w R

Figure 8. Contours for stream wise (u), velocity vertical velocity (v), lateral velocity (w), resultant velocity
(R) at 2.25m from head

Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020) 9

 Abinash Sahoo, Sandeep Samantaray and Rosysmita Bikram Singh

Figure 9 shows the velocity distribution contours across the flow cross-section at 0.03,
0.09, 0.15 m from the side walls simulated in ANSYS in no load condition.

(a)

(b)

(c)

Figure 9. Stream-wise velocity in lateral direction at (a) 0.03m, (b) 0.09m, (c) 0.15m from side wall

RESULTS AND ANALYSIS

Plot of stage vs discharge is represented with the data obtained from the experimental
observations performed in the laboratory as shown below in Figure 10. In the Horizontal
axis the value represents the discharge in cm3/sec and the Stage of the flow in m is indicated
in vertical axis in m (i.e. the height of flow).
The stream wise, vertical and lateral velocity components obtained from the
experimental set up are revealed in Table 2. The results (the resultant velocity) generated
from the experimental set up and the numerical analysis is indicated in Table 3. The overall

10 Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020)

 Analysis of Velocity Profiles in Rectangular Straight Open Channel Flow

flow patterns of experimental runs for the rectangular flume are summarized in the graphs.
In this paper the results for the test 5 i.e. at depth 0.22 m is shown. In similar way the results
of all the other different depths were found out and put for validation.

Figure 10. Plot of stage vs discharge

Table 2
Variation of stream wise velocity Vu, vertical velocity Vv, lateral velocity Vw

Stream wise velocity Vertical velocity Lateral velocity

Height Numerical Experimental Numerical Numerical
0.01 0.000915 0.00086 -0.00021 -0.50673
0.03625 0.001122 0.001043 -0.0003 -0.5469
0.0625 0.001305 0.001226 -0.00036 -0.60262
0.115 0.001405 0.001321 -0.0004 -0.63509
0.14125 0.00148 0.001391 0.000105 -0.67234
0.1675 0.001553 0.001459 0.00016 -0.72495
0.19375 0.001494 0.001394 0.000252 -0.72651
0.22 0.001391 0.001294 0.00028 -0.72718

Table 3
Variation of resultant velocity (Vr)
Height Numerical Height Numerical
0.005 0.1 0.05 0.585
0.015 0.2 0.0625 0.602618
0.022 0.281 0.08875 0.624936
0.027 0.37 0.115 0.666846
0.03 0.41 0.14125 0.711448
0.033 0.46 0.1675 0.730029
0.035 0.523 0.19375 0.712568
0.039 0.558 0.22 0.696472
0.044 0.57

Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020) 11

 Abinash Sahoo, Sandeep Samantaray and Rosysmita Bikram Singh

Graphs were plotted using the data extracted from ANSYS by placing plates at
different intervals for purpose of velocity profile analysis. The numerical data obtained
were compared to experimental data. Figure 11 shows the comparison of stream-wise
velocity profile between experimental and numerical results. From the present study, it
can be seen that magnitude of velocity is maximum at some point below free surface as in
case of narrow channels which is true for both the experimental and numerical simulation.
Variations of vertical velocity profile obtained from the numerical simulation are shown in
Figure 12. It is seen that magnitude of velocity gives negative values to nearly mid height
with respect to depth of flow and then changes to positive values of velocity (Song and Cox,
1993; Pittaluga & Imran, 2014). Figure 13 represents the variation of velocity in lateral
direction. The flow characteristics give all negative magnitude of velocity.

0.25 0.25

0.2 0.2
Depth (m)

0.15 0.15
Depth (m)

0.1 0.1

0.05 0.05

0 0
0 0.001 0.002 0.003 0 0.001 0.002 0.003
Velocity (Vu) Velocity (Vu)
Numerical Experimental Numerical Experimental
(a) (b)

0.25

0.2
Depth (m)

0.15

0.1

0.05

0
0 0.001 0.002 0.003
Velocity (Vu)
Numerical Experimental
(c)

Figure 11. Variation of stream wise velocity comparison for numerical and experimental result at (a) 0.25 m,
(b) 1.5m, (c) 2.25m from head gate

12 Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020)

 Analysis of Velocity Profiles in Rectangular Straight Open Channel Flow

0.25 0.25

0.2 0.2

0.15 0.15 Numerical

Depth (m)
Depth (m)

0.1 0.1
0.05 Numerical 0.05
0 0
-0.0004 -0.0002 0 0.0002 0.0004 -0.8 -0.6 -0.4 -0.2 0
Velocity (Vv) Velocity Vw
(a) (a)

0.25 0.25
0.2 0.2
0.15 Depth (m)
Depth (m)

0.15
Numerical
0.1 0.1
0.05 Numerical 0.05
0 0
-0.0003-0.0002-0.0001 0 0.0001 0.0002 0.0003 -0.8 -0.6 -0.4 -0.2 0
Velocity Vv Velocity (Vw)

(b) (b)

0.25 0.25
0.2 0.2
Depth (m)

0.15 0.15 Numerical

Depth (m)

0.1 0.1
Numerical
0.05 0.05

0 0
-0.0004 -0.0002 0 0.0002 0.0004 -0.8 -0.6 -0.4 -0.2 0
Velocity (Vv) Velocity (Vw)

(c) (c)

Figure 12. Variation of vertical velocity for numerical Figure 13. Variation of Lateral velocity for numerical
result at (a) 0.25 m, (b) 1.5m, (c) 2.25m from head gate result at (a) 0.25 m, (b) 1.5m, (c) 2.25m from head gate

The variation of resultant velocity profile found out by numerical simulation concerning
universal laws (Log and Power Law) is shown in Figure 14. It is observed that the
characteristics of flow profile below 0.2h (h-depth of flow) obeys Log law and above 0.2h
follows the Power law (Sarma et al., 1983; Castro-orgaz & Dey, 2011). The R2 values
from the resulting trend line show that the magnitude of velocity obtained best fit to the
universal laws. The equation of the two lines is given by ‘y’ which can be clearly indicated

Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020) 13

 Abinash Sahoo, Sandeep Samantaray and Rosysmita Bikram Singh

as the inner region follows a logarithmic path and the outer region follows an exponential
path. Figure 15 represents the comparison of stream-wise velocity between experimental
and numerical results at 0.03, 0.09 and 0.15m from side wall across the flow cross-section.

0.25

0.2 below 0.2h

above 0.2h y = 1.3156x5.9404
Log. (below 0.2h) R² = 0.8811
Depth (m)

0.15
Power ( above 0.2h)
0.1
y = 0.0207ln(x) + 0.0499
0.05 R² = 0.9518

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Velocity (Vr)

(a)

0.25 below 0.2h

above 0.2h
0.2 Log. (below 0.2h) y = 4.4481x9.4047
0.15 Power (above 0.2h) R² = 0.8972
Depth (m)

0.1 y = 0.0316ln(x) + 0.0561

0.05 R² = 0.9671

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Velocity (Vr)

(b)

0.25 below 0.2h

y = 2.0585x8.9546
0.2 above (0.2h)
R² = 0.9152
Log. (below 0.2h)
0.15
Depth (m)

Power (above (0.2h))

0.1
y = 0.0409ln(x) + 0.0594
0.05 R² = 0.9687

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Velocity (Vr)

(c)

Figure 14. Resultant velocity concerning to Log Law and Power Law at (a) 0.25m, (b) 1.25m, (c) 2.25m from
head gate

14 Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020)

 Analysis of Velocity Profiles in Rectangular Straight Open Channel Flow

0.25
Numerical
0.2
Experimental
Depth (m)

0.15

0.1

0.05

0
0 0.0002 0.0004 0.0006 0.0008 0.001 0.0012
Velocity (Vu)

(a)

0.25

0.2 Numerical
Experimental
0.15
Depth (m)

0.1

0.05

0
0 0.0002 0.0004 0.0006 0.0008 0.001 0.0012 0.0014
Velocity (Vu)

(b)

0.25

0.2 Numerical

Experimental
0.15
Depth (m)

0.1

0.05

0
0 0.0002 0.0004 0.0006 0.0008 0.001 0.0012 0.0014
Velocity (Vu)

(c)

Figure 15. Stream wise velocity comparison for numerical and experimental result at (a) 0.03m, (b) 0.09m,
(c) 0.15m from side wall

Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020) 15

 Abinash Sahoo, Sandeep Samantaray and Rosysmita Bikram Singh

Analysis
The velocity profile can be broken into two regions such as the inner region, below 0.2h and
the outer region, above 0.2h (where h=depth of flow). Velocity profile can be demonstrated
by the law of logarithmic distribution and with the help of a power law distribution up to
the point of maximum velocity. As per universal velocity profile laws, the region below
0.2h always follows logarithmic law and the region above 0.2h follows power law. The
experimental data and the numerical data extracted show that the velocity profile drawn
here fits well with the universal velocity profile laws.

CONCLUSION
In present scenario, the experimental and numerical investigation for prediction of velocity
for flow over plain bed is shown. In the primary stage of the present study, a 3D turbulence
model for flow over plain bed was simulated utilising k-ϵ closure model. With the help
experimental and numerical results, deviation in values for velocity of flow was presented
using contour mapping. The main distinguishing point between the velocity distribution
in laminar and turbulent flow conditions is that in case of flow to be laminar, the peak
velocity occurs at top surface of water while for most turbulent flow situations, the peak
velocity is found to be at some point below the water surface which is in case of an open
channel cross-section.
• It is examined that the peak velocity is not found at the water surface; rather it
occurs at some point below the top surface of water because of effect of turbulence
in the rectangular flume used for experiment during the fluid flow (Cardoso et al.,
1989).
• The velocity dip is visible from theexperimental results but in ANSYS inadequacy
of numerical modelling is observed which is often reasonable as the source of
computation are not worth to produce such better results.
• Additionally, at point when there is effect of sidewall, a lateral velocity component
(w) is harmonized close from sidewall of the channel to centre of channel and a
down flow (v) occurs from top water surface. The components of secondary flow
give the magnitude of maximum velocity below the top water surface. This is
observed for high depth of flow.
• Inner region of the flow depth is considered from bed of the channel to 0.2 of the
height of flow which follows the Logarithmic law, whereas from above 0.2 of
the flow height up to the top water surface is the outer region which follows the
power law.
• From the Figure 9 it can be found out that the numerical results are showing error
within 10% with respect to the experimental results. The best results are obtained

16 Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020)

 Analysis of Velocity Profiles in Rectangular Straight Open Channel Flow

for highest flow depth. For lowest flow depth, maximum deviation is observed
because of effect of boundary layer on the flow parameters.
• As it is observed from Figure 13, variation of velocity in vertical and lateral
direction, the numerical results could only be found out using ANSYS, because
of the use of Pitot tube which is usually a 1-D instrument and only measures the
stream-wise velocity.

ACKNOWLEDGEMENT
We thank the anonymous reviewers and editors for their useful suggestion to develop this
manuscript.

REFERENCE
Absi, R. (2011). An ordinary differential equation for velocity distribution and dip-phenomenon in open channel
flows. Journal of Hydraulic Research, 49(1), 82-89.

Afzal, N., Seena, A., & Bushra, A. (2007). Power law velocity profile in fully developed turbulent pipe and
channel flows. Journal of Hydraulic Engineering, 133(9), 1080-1086.

Bonakdari, H., Baghalian, S., Nazari, F., & Fazli, M. (2014). Numerical analysis and prediction of the velocity
field in curved open channel using artificial neural network and genetic algorithm. Journal of Engineering
Applications of Computational Fluid Mechanics, 5(3), 384-396.

Cardoso, A. H., Graf, W. H., & Gust, G. (1989). Uniform flow in a smooth open channel. Journal of Hydraulic
Research, 27(5), 603-616.

Castro-Orgaz, O., & Dey, S. (2011). Power-law velocity profile in turbulent boundary layers: An integral
Reynolds-number dependent solution. Acta Geophysica, 59(5), 993-1012.

Coles, D. (1956). The law of the wake in the turbulent boundary layer. Journal of Fluid Mechanics, 1(2),
191-226.

Guo, J., & Julien, P. Y. (2006, May 21-25). Application of modified log-wake law in open-channels. In World
Environmental and Water Resource Congress 2006: Examining the Confluence of Environmental and
Water Concerns (pp. 1-9). Omaha, Nebraska, United States.

Issa, R. I. (1986). Solution of the implicitly discretised fluid flow equations by operator-splitting. Journal of
Computational Physics, 62(1), 40-65.

Kang, H., & Choi, S. U. (2006). Reynolds stress modelling of rectangular open‐channel flow. International
Journal for Numerical Methods in Fluids, 51(11), 1319-1334.

Larocque, L. A., Imran, J., & Chaudhry, M. H., (2013). 3D numerical simulation of partial breach dam break
flow using the LES and k-𝜀 turbulence models. Journal of Hydraulic Research, 51(2), 145-157.

Naik, B., Khatua, K. K., Wright, N., Sleigh, A., & Singh, P. (2018). Numerical modeling of converging
compound channel flow. ISH Journal of Hydraulic Engineering, 24(3), 285-297.

Naot, D., Nezu, I., & Nakagawa, H. (1993). Hydrodynamic Behaviour of Compound Rectangular Open

Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020) 17

 Abinash Sahoo, Sandeep Samantaray and Rosysmita Bikram Singh

Channels. ASCE Journal of Hydraulic Engineering, 119(3), 390-408.

Nezu, I., & Rodi, W. (1986). Open-channel flow measurements with a laser Doppler anemometer. Journal of
Hydraulic Engineering, 112(5), 335-355.

Pittaluga, M. B., & Imran, J. (2014). A simple model for vertical profiles of velocity and suspended sediment
concentration in straight and curved submarine channels. Journal of Geophysical Research: Earth Surface,
119(3), 483-503.

Sarma, K. V. N., Lakshminarayana, P., & Rao, N. S. L. (1983). Velocity distribution in smooth rectangular
open channels. Journal of Hydraulics Engineering, 109(2), 270-289.

Sarma, N. V. K., Prasad, R. V. B., & Sarma, K. A. (2000). Detailed study of binary law for open channel.
Journal of Hydraulics Engineering, 126(3), 210-214.

Song, R., & Cox, S. K. (1993, June 14-17). Vertical velocity in cirrus case obtained from wind profiler. In
Proceedings of a Conference on the FIRE Cirrus Science (pp. 153-156). Breckenridge, Colorado, United
States.

Welahettige, P., Lie, B., & Vaagsaether, K. (2017). Flow regime changes at hydraulic jumps in an open Venturi
channel for Newtonian fluid. The Journal of Computational Multiphase Flows, 9(4), 169-179.

Wilkerson, V. G., & McGahan, J. L. (2005). Depth averaged velocity distribution in straight trapezoidal channel.
Journal of Hydraulics Engineering, 131(6), 509-512.

18 Pertanika J. Sci. & Technol. 28 (1): 1 - 18 (2020)