NUMERICAL AND EXPERIMENTAL INVESTIGATION OF RIVER HYDROKINETIC

TURBINE FOR WATER PUMPING APPLICATION

Muzammil Ejaz1, Mohammad Fozan ur Rab2, Fahad Qureshi4, Hassan5

NED University of Engineering and Technology, Karachi, Pakistan

Wajiha Rehman3
University of Engineering and Technology Lahore, Lahore, Pakistan

ABSTRACT
This study is carried out to design a low head turbine which will utilize the kinetic energy of
flowing river stream. This turbine will be used to drive a pump by coupling the turbine’s shaft with
the pump’s shaft. As the turbine rotates, the pump also rotates and pumps the water to the desired
altitude where water is needed by a remote community. In this way, this system requires no
external power and heavy construction work. Axial flow Hydrokinetic turbine’s design is chosen
because it has high efficiency and fabrication ease. In this paper, a scaled down prototype of actual
turbine is fabricated and tested in a water tunnel at 0.7 m/sec flow velocity. Based on experimental
work, a CFD model is designed. The turbine’s prototype model is analyzed by using
Computational Fluid Dynamics (CFD) tool at same inlet conditions and then numerical value of
moment is compared with the experimental value of moment at same rotational speed. Both values
of moments are very close to each other which validates the CFD model. The k-omega SST
turbulence model is utilized. The computational grid contains approximately 5 million cells with
average y+ value less than five. In future, this validated CFD model will be used to simulate the
full-scale turbine prototype in actual river conditions and will be helpful in optimizing the turbine’s
model.

INTRODUCTION
Renewable resources are capable of meeting energy demands while decreasing the effect of global
warming1. Water is a dominant renewable energy resource that is supplying 18% of global
electricity. The approximated global gross, economic, technical and exploitable hydropower
potential is estimated as 128, 21, 26 and 16 Penta watt hours per year, respectively2. According to
the NEPRA report3, Pakistan’s hydel potential is 40,000 MW but only 15% of this potential is
exploited. A huge potential is available in the form of low head e.g. canals, run of rivers and farm
channels where conventional turbine systems cannot be used. Hence, a novel technology must be
used to exploit this potential.

Currently, the two major methods to produce electricity from water are hydrostatic method and
hydrokinetic method. In hydrostatic method, dams and other civil structures are used to store water
and then its energy is extracted by turbines4. In hydrokinetic system, the kinetic energy of the
moving fluid is converted into electricity5. Globally, various studies have done to examine the
working and designing parameters of hydrokinetic turbine by using various approaches.
Hydrokinetic turbine system is a popular research topic, which intrigued engineers from all over

TFAWS 2018 – August 20-24, 2018 1

the world. Numerical tools like Computational Fluid Dynamics (CFD) is one of the most powerful
tools that has revolutionized the designing procedure. Many researchers, from all over the world,
are using CFD to design hydrokinetic turbines. Kolekar et al.6-8 and Mukherji et al.9 optimized the
design of a 12 kW-horizontal axis turbine by employing CFD and blade element theory. Batten et
al.10-12 used Blade Element Momentum model (BEM) to design a horizontal axis turbine for the
exploitation of tidal energy. They also performed experiments in a cavitation tunnel by using a
scaled model and found that BEM model satisfied experimental results. Researchers13-15 improved
BEM model to consider the effects of dynamic stall, rotational flow and losses at the tip of blade.
Later, three-dimensional inviscid models were introduced to explain the hydrodynamics of turbine
in more detail than the BEM model. Researchers16-18 improved 3D inviscid model by introducing
lifting line, panel and vortex lattice method but it was incapable of taking the effect of viscous
forces that are required for accurate performance prediction. Tian et al.19 designed and performed
CDF analysis of a Horizontal Axis Water Turbine (HAWT) to power the Underwater Moored
Platforms (UMP). They used ANSYS FLUENT v13.0 as CFD tool and selected k-ω Shear Stress
transport (SST) model to solve the Reynold’s Averaged Navier Stokes (RANS) equations. The
results of the simulation were validated by experimental data and which supported their numerical
model. Lawson et al.20 optimized the design of a Horizontal Axis Tidal Turbine (HATT) and
employed k-omega SST turbulence model to get results for the turbine’s performance. The results
obtained were compared with BEM and there was a reasonable agreement. Nak et al.21 also
designed a HATT and performed experimental and CFD analysis by using ANSYS CFX v13.0.
Authors used k-ω SST turbulence model and plot the performance characteristics curves of the
rotor and found the CFD results quite accurate. Various other researchers22-24 used k-omega SST
turbulence model for the prediction of turbine’s performance and found good agreement with the
experimental data.

Mayer and Bahaj25 performed experiments on horizontal axis water turbine to determine the
designing parameters of rotor. They found that the stall delay of the foil greatly depends on the
twist angle of blade, lift and drag performance, centrifugal force and rotor’s yaw angle. Hayati et
al.26 found that the thrust performance of turbine increased by increasing the rake angle of the
propeller. Singh and Nestmann27, 28 did parametric study to optimize the geometry of rotor. By
modifying the inlet tip of the runner to reduce discharge consumption, they improved efficiency
i.e. from 55% to 74%. They modified the inlet and exit angles of the blades, for three different
stages of runner, and found that was increased by decreasing the exit blade angle. Singh and
Nestmann26 found designing parameters of low head horizontal axis water turbine. They found the
relation of blade number with turbine’s performance and concluded that blade number is more
significant than blade’s height to achieve high efficiency.

In this paper, a low head turbine is designed to power a pump which will pump water from river
to houses in Gilgit, Pakistan. The residential area is at the elevation of 150 ft. from the river. The
proposed site of reservoir tank is 1702 ft. away from river while the minimum depth of river is 5
ft. (depending on summer and winter). The specifications of the site are mentioned in Fig. 1.

TFAWS 2018 – August 20-24, 2018 2

 Figure 1. Aerial map of Gilgit Baltistan site
It is investigated that hydrokinetic turbine is feasible for installation on site. The turbine would
be installed on a floating platform with steel cable-ties support. The turbine will be fully
submerged and pump will be coupled with turbine’s shaft. The arrangement of turbine is shown
in Figure 2.

TFAWS 2018 – August 20-24, 2018 3

 Figure 2. Conceptual model of complete turbine assembly

TURBINE DESIGN
In this paper, the design of runner is taken from Schleicher et al. 30 works. The turbine’s rotor and
diffuser geometry with solid cylindrical outer zone for Computational Fluid Dynamics (CFD)
analysis was prepared by using Computer-Aided Design (CAD) software. The Tip and Hub
diameters of turbine are Dt=0.0844 m and Dh= 0.0201 m respectively while the diffuser inlet and
outlet diameters are 0.1025 and 0.1118 m.

Figure 3. (a) Front view of rotor, (b) Side view of rotor (c) Side view of diffuser

TFAWS 2018 – August 20-24, 2018 4

 Figure 4. CAD model of rotor with casing

EXPERIMENTAL ANALYSIS
Experimental setup

At first the experimental setup was designed to test the scaled model of turbine. So that, on the
basis of experimental results, a Computational Fluid Dynamics (CFD) model can be designed and
its performance can be validated. The scaled prototype was 3D printed and then tested in water
tunnel in Fluid Mechanics laboratory. The turbine and diffuser were printed by using Polylactic
Acid (PLA) plastic material and then coated with synthetic paints to get smooth surface. PLA
plastic is different from thermoplastic polymers because it is derived from renewable resources
like cornstarch or sugar cane. The turbine was placed in a water tunnel. Width of the testing
channel was 30 cm while water height varied depending upon the flow rate of water. At a flow
rate of 23 L/sec the height of water in channel was around 10.5 cm and the turbine was fully
submerged in water. The velocity of water was 0.73 m/s. Under these conditions, rotational speed
of turbine was measured using a slow-motion video camera, turbine was rotating around 480 RPM
without any external braking force (free run).

To calculate the power utilized by turbine at 450 rpms, an electrical Direct Current (DC) motor
was used which was assumed to be 80% efficient. Turbine shaft was attached to the DC motor and
DC was supplied. The rpms were calculated by using a laser gun tachometer. At a voltage of 5.1
V and 0.1 Amp current, the turbine was running at 480 rpms. Using the relation in equation number
7, 0.5 watts of electrical power was obtained. As DC motor was 80% efficient, the mechanical
power was calculated to be around 0.408 watts. Using the relations given in Eq. 8 and Eq. 9, the
torque generated by turbine at 480 rpms was calculated.

𝑃 =𝐼∗𝑉 (7)

TFAWS 2018 – August 20-24, 2018 5

 0.408 = (0.1 ∗ 5.1) ∗ 0.8

𝑃=𝑇∗ 𝜔 (8)

2𝜋𝑁𝑇 (9)
𝑃=
60
Here, P is power, I is current in amps, V is voltage, T is torque, ω is angular velocity and N is
revolutions per minute.

Table 2. Applied conditions on the test bench

Input electrical power Shaft power Speed Torque applied
(Electrical power* 0.8)

Watts Watts RPM N-m

0.51 0.408 480 0.00812

Figure 5. Experimental setup of turbine

Results of experimental analysis

TFAWS 2018 – August 20-24, 2018 6

 Table 3. Results obtained under applied conditions
Flow rate Speed at free run Speed at applied Velocity
torque

Liter/sec RPM RPM m/sec

23 435-450 405-413 0.6-0.7

NUMERICAL ANALYSIS
Computational Grid and Boundary Conditions

Rotating reference frame was used in the proximity of turbine to transform flow from inertial frame
of reference to non-inertial frame of reference to capture rotational effects. The diameter of rotating
reference frame was kept slightly larger than Dt and its length was also greater than the length of
diffuser. Hexahedral dominant grid was utilized to discretize the flow domain into approximately
5 million cells. The grid that was generated comprised of varying degree of refinement levels in
different regions of flow domain. The generated grid resulted in average y+ value ranging from 0
to 5 along the walls of turbine and diffuser.

Figure 6. Meshed model of turbine

TFAWS 2018 – August 20-24, 2018 7

 Figure 7. Meshed model of turbine and flow field area

The length of the computational domain is 2.5 m while both height and width are 1 m each. The
computations were performed in SimScale (Open FOAM based Cloud simulation platform) using
pressure-based segregated Semi-Implicit Method for Pressure-Linked Equations (SIMPLE)
algorithm. The uniform velocity inlet boundary condition of 0.7 m/sec was specified at the inlet of
flow domain. The constant gauge pressure of 0 Pa was set at the domain outlet. Full resolution
wall treatment was applied to both the walls of turbine and diffuser section. The lower face of
domain was set to no-slip wall while the remaining faces were modeled as slip-wall with zero
velocity gradient. The moving reference frame angular velocity was set to 47 radian/sec
corresponding to rotational speed of scaled model in experiment.

TFAWS 2018 – August 20-24, 2018 8

 Figure 8. Flow field with labelled boundary conditions

Governing equations
The flow is assumed incompressible, turbulent and steady. All thermo-physical properties of water
are independent of temperature. Governing equations in cylindrical coordinates are written as,

Mass conservation:

1 (ur ) 1 (u ) (uz ) (1)

  0
r r r  z

Momentum conservation:

r -component:

 (ur )  (ur ) u  (ur ) u2 ur

(  ur    uz )
t r r  r z
P 1  ur ur 1  2ur 2 u  2ur
   g r  [ (r ) 2  2  2  ] (2)
r r r r r r  2 r  z 2
θ -component:

 (u )  (u ) u  (u ) u2 u

(  ur    uz  )
t r r  r z
1 P 1  u u 1  2u 2 ur  2u
   g  [ (r ) 2  2  2  ] (3)
r r r r r r r  2 r  z 2
z -component:

TFAWS 2018 – August 20-24, 2018 9

  (u z )  (u z ) u  (u z ) u z
(  ur    uz )
t r r  z
P 1  u z 1  2u z  2u r
   g z  [ (r ) 2  ] (4)
r r r r r  2 z 2
In equations, 𝑢𝑟 , 𝑢𝜃 , 𝑢𝑧 are radial, tangential and axial components of velocity while 𝑔𝑟 , 𝑔𝜃 , 𝑔𝑧 are
radial, tangential and axial components of gravitational acceleration, respectively. The density of
water is denoted by ρ while dynamic viscosity is denoted by µ.

Turbulence Modelling

Menter’s k-omega Shear Stress Transport (SST) two-equation eddy-viscosity model was used to
model turbulence. The k-ω SST can model the transport of turbulent shear stress while accurately
predicting the onset and amount of flow separation under adverse pressure gradient. It
demonstrates k-epsilon type behavior in far field and is better at capturing free-shear effects while
it switches to k-omega formulation in the near-wall region inside boundary layer. k-epsilon model
is usually not suitable for modelling large adverse pressure gradients while k-omega has its
limitations in modelling free-shear flows (wake, recirculation). k-omega SST model incorporates
Bradshaw’s 31 observation that the principal turbulent shear stress is proportional to the turbulence
kinetic energy in the wake region of boundary layer by redefining the eddy viscosity formulation
for adverse pressure gradient boundary layer. Hence, it is the most suitable model for the
simulation of turbine. The equations for turbulence kinetic energy and specific dissipation rate are
written as

( pk )   k
 ( pkui )  ( k )  Gk  Yk
t xi x j x j (5)

( p )   
 ( pui )  ( )  G  Y  D
t xi x j x j (6)

Here, 𝛤𝑘 and Γ𝜔 represent the effective diffusivity for k and ω. The turbulence kinetic energy
generated by velocity gradient is represented by 𝐺𝑘 while 𝐺𝜔 represents the generation of ω. The
dissipation in k and ω, due to turbulence, is represented by 𝑌𝑘 and𝑌𝜔 , respectively. 𝐷𝜔 shows the
cross-diffusion term.

Results of numerical analysis

The computations were carried out in SimScale (OpenFOAM backed Cloud based simulation
platform) using steady-state pressure-based segregated Semi-Implicit Method for Pressure-Linked
Equations (SIMPLE) algorithm. Pressure moment in z-direction is found to be 0.00760336 N-m
which is in good agreement with experimental data.

TFAWS 2018 – August 20-24, 2018 10

 Figure 9. Flow velocity streamlines and turbine surface pressure contours
In figure 9, the streamlines of flow colored by velocity magnitude are plotted and it has been
observed that the flow after leaving the turbine becomes highly rotational in nature. Both turbine
and diffuser surfaces are also colored by surface pressure contours.

Table 1. Results obtained from Numerical Analysis

Upstream Rotational Torque Mechanical Power Tip-
flow velocity speed power coefficient Speed
Ratio

m/sec RPM N-m Watts

0.7 480 0.0076033 0.3801676 0.39728056 3.014285

TFAWS 2018 – August 20-24, 2018 11

 Figure 10. Variation of pressure moment along rotational axis with simulation time
Figure 10 represent trend of pressure moment along rotational axis with the evolution of
simulation time. As the simulation proceeds, the large variations in moment value are
diminished.

TFAWS 2018 – August 20-24, 2018 12

 Figure 11. Contours of static pressure
Figure 11 illustrates that the pressure is greatest near the shaft cone of turbine and it is major
contributor to the thrust force experienced by shrouded turbine.

TFAWS 2018 – August 20-24, 2018 13

 Fig.12 Contours of Velocity magnitude

In figure 12, the velocity contours plot depicts that the region behind the turbine is dominated
by strong wake. Small vortices are shed by the diffuser and the diffuser section is also
contributing to increase in flow velocity near blades. The presence of diffuser results in improved
efficiency of turbine.

TFAWS 2018 – August 20-24, 2018 14

 0.5
0.45
0.4
Power Coefficient (Cp) 0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
0 0.5 1 1.5 2 2.5 3 3.5
Tip-Speed Ratio (λ)

Figure 13. Relation of Cp vs. TSR

The graph was plotted between the power coefficient and tip speed ratio in figure 13. The highest
efficiency was achieved at TSR 2.4 which is 44%. By increasing the TSR beyond 2.4 and a
decreased in efficiency was observed.

1.4

1.2
Thrust coefficientCT)

0.8

0.6

0.4

0.2

0
0 0.5 1 1.5 2 2.5 3 3.5
Tip-Speed Ratio(λ)

Figure 14. Relation of CT with TSR

In Figure 14, a graph was plotted between thrust coefficient and tip speed ratio to study the
relationship of thrust coefficient with TSR. The highest CT value was obtained at TSR of
approximately 1 and decrease in CT was observed with the increase in TSR.

TFAWS 2018 – August 20-24, 2018 15

CONCLUSIONS
The performance of Hydrokinetic turbine is quantified both experimentally and numerically. The
best efficiency point is obtained at TSR of about 2.4. The results obtained from CFD are in good
agreement with experimental values having deviation of about 6.3633 %. Both numerical analysis
and experimental investigation of scaled model of axial hydro-kinetic turbine conducted in
university laboratory have shown very promising trends for development of full-scale prototype.

ACKNOWLEDGEMENTS
The authors are grateful to Mr. Ghulam Farooq (Senior Executive, Pak Suzuki Motor Company
Ltd.) and Mr. Muhammad Affan (Student, Department of Electrical Engineering, NED University
of Engineering and Technology) for their unconditional help and support.

REFERENCES
1. Müller, B., Arlt, W., & Wasserscheid, P. (2011). A new concept for the global distribution
of solar energy: energy carrying compounds. Energy & Environmental Science, 4(10),
4322-4331.

2. Zhou, Y., Hejazi, M., Smith, S., Edmonds, J., Li, H., Clarke, L., & Thomson, A. (2015). A
comprehensive view of global potential for hydro-generated electricity. Energy &
Environmental Science, 8(9), 2622-2633.

3. NEPRA. (2018, 7 1). Hydel Potential in Pakistan. Retrieved from NEPRA:

http://www.nepra.org.pk/Policies/Hydel%20Potential%20in%20Pakistan.pdf

4. Khan, M. J., Iqbal, M. T., & Quaicoe, J. E. (2008). River current energy conversion
systems: Progress, prospects and challenges. Renewable and Sustainable Energy
Reviews, 12(8), 2177-2193
5. Güney, M. S., & Kaygusuz, K. (2010). Hydrokinetic energy conversion systems: A
technology status review. Renewable and Sustainable Energy Reviews, 14(9), 2996-3004

6. Kolekar, N., Mukherji, S. S., & Banerjee, A. (2011, January). Numerical modeling and
optimization of hydrokinetic turbine. In ASME 2011 5th International Conference on
Energy Sustainability (pp. 1211-1218). American Society of Mechanical Engineers.

7. Kolekar, N., Hu, Z., Banerjee, A., & Du, X. (2013, April). Hydrodynamic design and
optimization of hydro-kinetic turbines using a robust design method. In Proceedings of the
1st Marine Energy Technology Symposium.

8. Kolekar, N., & Banerjee, A. (2015). Performance characterization and placement of a

marine hydrokinetic turbine in a tidal channel under boundary proximity and blockage
effects. Applied Energy, 148, 121-133.

TFAWS 2018 – August 20-24, 2018 16

9. Subhra Mukherji, S., Kolekar, N., Banerjee, A., & Mishra, R. (2011). Numerical
investigation and evaluation of optimum hydrodynamic performance of a horizontal axis
hydrokinetic turbine. Journal of Renewable and Sustainable energy, 3(6), 063105.

10. Batten, W. M. J., Bahaj, A. S., Molland, A. F., & Chaplin, J. R. (2006). Hydrodynamics of
marine current turbines. Renewable energy, 31(2), 249-256.

11. Batten, W. M. J., Bahaj, A. S., Molland, A. F., & Chaplin, J. R. (2008). The prediction of
the hydrodynamic performance of marine current turbines. Renewable energy, 33(5),
1085-1096.

12. Batten, W. M. J., Bahaj, A. S., Molland, A. F., Chaplin, J. R., & Sustainable Energy
Research Group. (2007). Experimentally validated numerical method for the
hydrodynamic design of horizontal axis tidal turbines. Ocean engineering, 34(7), 1013-
1020.

13. Leishman, J. G., & Beddoes, T. S. (1989). A Semi‐ Empirical model for dynamic
stall. Journal of the American Helicopter society, 34(3), 3-17.

14. Burton, T., Sharpe, D., & Jenkins, N. (2001). Handbook of wind energy. John Wiley &
Sons.

15. Shen, W. Z., Mikkelsen, R., Sørensen, J. N., & Bak, C. (2005). Tip loss corrections for
wind turbine computations. Wind Energy: An International Journal for Progress and
Applications in Wind Power Conversion Technology, 8(4), 457-475.

16. P. Epps, J. Stanway, W. Kimball (2009). An open-source design tool for propellers and
turbines. Society of Naval Architects and Marine Engineers Propeller/Shafting '09
Symposium, Williamsburg, VA

17. Liu, P. (2010). A computational hydrodynamics method for horizontal axis turbine–Panel
method modeling migration from propulsion to turbine energy. Energy, 35(7), 2843-2851.

18. He, L., Kinnas, S. A., & Xu, W. (2013). A Vortex-Lattice Method for the Prediction of
Unsteady Performance of Marine Propellers and Current Turbines. International Journal
of Offshore and Polar Engineering, 23(03).

19. Myers, L., & Bahaj, A. S. (2006). Power output performance characteristics of a horizontal
axis marine current turbine. Renewable energy, 31(2), 197-208.

20. Lawson, M. J., Li, Y., & Sale, D. C. (2011, January). Development and verification of a
computational fluid dynamics model of a horizontal-axis tidal current turbine. In ASME
2011 30th International Conference on Ocean, Offshore and Arctic Engineering (pp. 711-
720). American Society of Mechanical Engineers.

21. Nak-Joong, L., In-Chul, K., Beom-Soo, H., & Young-Ho, L. (2015). Experimental and
Numerical Investigation of Blade Angle Variation on a Counter-Rotating Tidal Current

TFAWS 2018 – August 20-24, 2018 17

 Turbine. In Renewable Energy in the Service of Mankind Vol I(pp. 305-316). Springer,
Cham.

22. Pape, A. L., & Lecanu, J. (2004). 3D Navier–Stokes computations of a stall‐ regulated
wind turbine. Wind Energy: An International Journal for Progress and Applications in
Wind Power Conversion Technology, 7(4), 309-324.

23. Tongchitpakdee, C., Benjanirat, S., & Sankar, L. N. (2005). Numerical simulation of the
aerodynamics of horizontal axis wind turbines under yawed flow conditions. Journal of
solar energy engineering, 127(4), 464-474.

24. Sørensen, N. N., Michelsen, J. A., & Schreck, S. (2002). Navier–Stokes predictions of the
NREL phase VI rotor in the NASA Ames 80 ft× 120 ft wind tunnel. Wind Energy: An
International Journal for Progress and Applications in Wind Power Conversion
Technology, 5(2‐ 3), 151-169.

25. Tian, W., Song, B., VanZwieten, J. H., Pyakurel, P., & Li, Y. (2016). Numerical
simulations of a horizontal axis water turbine designed for underwater mooring
platforms. International Journal of Naval Architecture and Ocean Engineering, 8(1), 73-
82.

26. Hayati, A. N., Hashemi, S. M., & Shams, M. (2012). A study on the effect of the rake angle
on the performance of marine propellers. Proceedings of the Institution of Mechanical
Engineers, Part C: Journal of Mechanical Engineering Science, 226(4), 940-955.

27. Singh, P., & Nestmann, F. (2011). Experimental investigation of the influence of blade
height and blade number on the performance of low head axial flow turbines. Renewable
Energy, 36(1), 272-281.

28. Singh, P., & Nestmann, F. (2010). Exit blade geometry and part-load performance of small
axial flow propeller turbines: An experimental investigation. Experimental Thermal and
Fluid Science, 34(6), 798-811.

29. Singh, P., & Nestmann, F. (2009). Experimental optimization of a free vortex propeller
runner for micro hydro application. Experimental Thermal and Fluid Science, 33(6), 991-
1002.

30. Schleicher, W. C., Riglin, J. D., & Oztekin, A. (2015). Numerical characterization of a
preliminary portable micro-hydrokinetic turbine rotor design. Renewable Energy, 76, 234-
241.

31. Menter, F. R. (1994). Two-equation eddy-viscosity turbulence models for engineering

applications. AIAA journal, 32(8), 1598-1605

TFAWS 2018 – August 20-24, 2018 18