Design of Water Pumping System By Wind Turbine For Using in Coastal

Areas of Bangladesh

Dr. Md. Mahbubul Alam, Tamanna Tasnim, Umnia Dohaa)

Department of Mechanical Engineering, Bangladesh University of Engineering

and Technology, Dhaka-1000

ABSTRACT
In this work, a theoretical analysis has been carried out to analyze the prospect of
Wind Pumping System (WPS) for using in coastal areas of Bangladesh. Wind
speed data of three coastal areas of Bangladesh- Kutubdia, Patenga and Sathkhira
has been analyzed and an optimal wind turbine viable for this wind speed range
has been designed using the simulation software Q-blade. The simulated turbine is
then coupled with a rotodynamic pump. The output of the Wind Pumping System
(WPS) for the three coastal areas has been studied.

a)Corresponding authors email address: umnia921@gmail.com

1. Introduction
Energy consumption is the prime indicator of development and economic growth
of any nation. With the booming economic activity growth of the developing
countries and the ongoing global development process, the need for energy sources
is mandatory. Unfortunately, the conventional energy sources- primarily fossil fuel
is limited and is expected to deploy by the end of this century. Therefore it is
essential to search for alternative energy sources in order to meet the ever-growing
demand of energy. Various research is going on to develop alternative energy
sources. Of them, the nuclear energy and renewable energy is considered to have
the most potential to supplant conventional fossil fuel. Renewable energy can be
defined as the energy which is naturally replenished on a human timescale.
Renewable energy replaces conventional fuels in different distinct areas: electricity
generation, pumping, air and water heating/cooling, motor fuels, and rural (off-grid)
energy supply. Being a natural source, renewable energy is free of cost and will be
cost effective if the initial cost of construction can be reduced by proper design of
energy harvesting system. Another benefit of renewable energy is its minimum
environmental impact.
Various types of renewable energy sources are currently being researched across
the globe and high throughput devices and technologies have been developed over
the centuries to harness renewable energy. Of them, the most common form of
renewable energy sources are solar, hydro, wind, geothermal, biomass, geothermal
and tidal energy. The practice of harnessing wind energy dates back to around 200
B.C. In modern days, with technological and structural advancement, wind energy
is currently considered a promising source of clean energy. The most common type
of wind energy-harnessing machine is the wind turbine. Two types of turbines are
in use for this purpose- Horizontal-axis Wind Turbine (HAWT), which are lift-
devices and Vertical-axis Wind Turbine (VAWT), which are drag devices. In our
study, we designed a HAWT as it has the advantage of obtaining a higher tip speed
than the wind speed and consequently, a higher power output to weight ratio[1].
Since Bangladesh is also one of the most densely populated countries in the world,
the consequence of failure providing proper safety features in case of nuclear
power generation will be disastrous. On the other hand, renewable energy is free of
causing any environmental or health hazard. So development of reliable and
consistent renewable technology is essential for sustainable development of
Bangladesh. With this regard, the study of the prospect of wind potential and the
feasibility study of wind power technology in Bangladesh using innovative
techniques is a very promising and attractive feature to meet the energy demand.
The wind speed data of Bangladesh reveals that Bangladesh has insignificant wind
speed for generating power[2], but it is sufficient in some coastal areas for drawing
water if it is coupled with a pump. So, the aim of this study is to design an
optimum Wind Pumping System (WPS) for coastal areas of Bangladesh so that
wind energy can be harnessed properly. Having an agriculture-based economy, a
significant amount of energy is consumed in irrigation. So, an efficient WPS can
also reduce a huge energy load from the conventional source.
The WPS is designed by coupling a horizontal-axis wind turbine with a centrifugal
pump. An optimum wind turbine suitable for the three selected locations- Kutubdia,
Potenga and Sathkhira has been designed using simulation software Q-blade[4]
and the performance of the WPS for the three locations has been evaluated in this
paper.

2. Theory and Operating Principle

2.1 Power Available in the Wind
The energy available from a wind turbine can be expressed by the following
equation:
1
KE ATv 3
2 ..(2.1)
Where, KE is the kinetic energy available from the wind turbine, is the density of
the wind, A is the area swept by the turbine blades, T is the time required for a
parcel of air with thickness D to move through the wind turbine, v is the speed of
the wind
The power available from the turbine can be obtained by dividing equation (2.1) by
time T, which yields,
1
P Av 3
2 (2.2)
2.2 Power Coefficient of the Wind Turbine
The coefficient of power is the most important variable in wind turbine
aerodynamics. Buckingham theorem can be applied to show that non-
dimensional variable for power is given by the equation below. This equation is
similar to efficiency, so values between 0 and less than one are typical. However
this is not the exactly the same as efficiency so in practice some turbines can
exhibit greater than unity power coefficients. In these circumstances one cannot
conclude the first law of thermodynamics is violated because this is not an
efficiency term by the strict definition of efficiency.
 P
C p (2.3)
1 3
Av
2
Where, Cp is the coefficient of power, P is the power available in the wind turbine,
is the density of air, A is the area swept by the turbine blades and v is the
velocity of the wind.
2.3 Tip Speed Ratio
When a rotor blade passes through the air it leaves turbulence in its wake. If the
next blade on the spinning rotor arrives at this point while the air is still turbulent,
it will not be able to extract power efficiently from the wind. However if the rotor
span a little more slowly the air hitting each turbine blade would no longer be
turbulent. Therefore the tip speed ratio is chosen so that the blades do not pass
through turbulent air.
Tip speed of blades
Tip Speed Ratio
wind speed .(2.4)
The optimum tip speed ratio depends on the number of blades in the wind turbine
rotor. The fewer the number of blades, the faster the wind turbine rotor needs to
turn to extract maximum power from the wind. A two-bladed rotor has an optimum
tip speed ratio of around 6, a three-bladed rotor around 5, and a four-bladed rotor
around 3 [6].
2.4 Specific Speed of a Centrifugal Pump
Specific speed of a centrifugal pump may be defined as the speed in revolutions
per minute of a pump geometrically similar to the actual pump, which will
discharge unit quantity of water (1m3/s), while it is being raised through a unit
head (1 meter). The performance and dimensional proportions of pumps having the
same specific speed will be the same even though their outside diameters and
actual operating speed may be different.[5]
Specific speed of a centrifugal pump is given by the following equation:

n Q
no (2.5)
H 3/ 4
Where, no is the specific speed of the pump, n is the nominal speed of the pump, Q
is the discharge and H is the head.
2.6 Optimum Gear Ratio
Gasch[3] developed the following equation for the optimum gear ratio of a WPS:
1/ 3
5
N

2n n 3 D

3 C p ( opt ) 2 5/3

i 2 2R
( opt ) Pn D N .(2.6)

Where, i is the optimum gear ratio, is the density of air, 1.225 kg/m3, opt is the
optimum tip speed ratio, Cp is the optimum power of coefficient of the turbine, R is
the rotor radius of the turbine, nn, Dn, Pn are the nominal speed, diameter and
power of the pump respectively.

3. Design Methodology
3.1 Designing Optimum Wind Turbine
For designing an optimum wind turbine for the WPS, an open source simulation
software Qblade[4] is used. Qblade uses the Blade Element Momentum (BEM)
theory[7] for simulating optimum blade design and turbine performance.
In this study, a NACA 23012 airfoil was chosen for the turbine blade. For the
simulation process, the following turbine parameters have been selected:
Rotor diameter, R = 5m
Hub height of the turbine, h = 30m
Number of blades = 3
For simulating the turbines variation of coefficient of performance over a range of
tip speed ratio, all the simulation parameters were specified and the dimensionless
simulation process was conducted for a range of tip speed ratio from 0 to 12 for
each of the blade profile. Thus the performance curve (power coefficient, Cp vs.
tip speed ratio, TSR) for the turbine was obtained.
Fig 3.1 (a) designed NACA 23012 aerofoil turbine blade profile and (b) the
characteristic curve of the wind turbine using Qblade

3.2 Selection of the Centrifugal Pump

A rotodynamic pump (centrifugal pump) has a greater advantage over a positive
displacement pump for Wind Pumping Systems because of its lower starting
torque and higher efficiency. A commercial radial centrifugal pump with the
following specifications was selected:
Specific speed: 27
Impeller diameter: 180 mm
Pump head: 42 m
Nominal speed: 2900 rpm
Nominal power: 15kW
 Maximum efficiency: 83.5%
For pumping water for irrigation purposes, such a large pump with high head is not
required. So applying the principle of specific speed[5], a pump model with a head
of 6m is constructed using equation (2.5). The nominal speed of the model is 1750
rpm. The nominal power of the pump model is obtained using the following
similarity equation:

P1 n13
(3.1)
P2 n23

Where, P1 and n1 are the nominal power and nominal speed of the model and P2
and n2 are the nominal power and nominal speed of the prototype respectively.
Putting n1= 1750 rpm, n2= 2900 rpm, P2= 15 kW, the nominal power of the model
pump is found to be P1=3300W.
The optimum gear ratio for the system is obtained by putting nn=n1=1750 rpm,
Pn=P1=3300W, D=180mm, R=2.5m, =1.225kg/m3, opt= 4.19 for 3-blade wind
turbine[6] and Cp= 0.3887 from the characteristic curve obtained from Fig 3.1b in
equation (2.6). This is found to be equal to 2.36.

3.3 Coupling of the Wind Turbine and the Pump

The wind mapping data of Bangladesh[2] reveals that the coastal areas of
Bangladesh has a higher wind potential than other regions of the country. So in this
study, three coastal areas of Bangladesh- Kutubdia, Patenga and Sathkhira was
chosen for designing the WPS.
Locati Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec
ons

Kutub 4.7 4.8 5.59 7.53 9.48 9.75 8.21 7.79 7.32 6.42 5.88 4.92
dia

Pateng 6.22 6.34 7.37 7.92 8.47 8.69 9.20 8.54 7.48 6.93 6.71 5.91
a

Sathkh 4.21 4.40 4.84 7.1 6.11 4.76 4.27 4.03 3.62 3.78 3.54 2.81
ira
Table 3.1 monthly average wind speed (m/s) of Kutubdia, Patenga and Sathkhira at
20 meters height [2]
The coupling of the simulated wind turbine and the selected pump is done in the
following steps:
Obtaining the power vs. wind speed curve of the turbine for the three locations
from the simulated characteristic curve of the turbine
Obtaining the power vs. speed curve of the WPS by plotting the power vs.
speed curve of the pump to same graph of the turbine
Finding the output power of the WPS from the combined power vs. speed
curve and calculating the discharge and overall efficiency from of the system
From the characteristic curve of the simulated wind turbine (Fig 3.1b), the values
of the tip speed ratio and the corresponding power coefficient Cp is determined
using the software Plot Digitizer[8]. Then putting the values of tip speed ratio and
power coefficient in equation (2.4) and (2.3) respectively, for various wind speeds
at the three selected locations, the power vs. annual average wind speed of the
turbine is obtained.
Fig 3.2 Power vs rotational speed curve of the designed wind turbine model at
monthly average wind speeds for the three locations (a) Kutubdia (b) Patenga and
(c)Sathkhira
The equation of the gear ratio can be expressed as,
n
i (3.2)
nt

Where, i is the gear ratio,n is the speed of the pump, nt is the speed of the turbine
The power vs. speed curve of the WPS can be found by the following steps:
Putting i =2.36 and different values for nt from the turbine power vs. rpm
curve of the turbine in equation (3.2), different rotational speed, n of the pump
is found
Putting n= n, n= 1750rpm and P= 3300W in equation (3.1) different values
of P corresponding to n= n is found
P vs. n is plotted in the same graph of the turbines power vs. rpm curve.
Thus the power vs. speed curve of the combined WPS system is obtained
The output power of the WPS is next obtained graphically from the intercept points
of the power vs. rpm curves of the turbine and the pump.
Fig 3.3 Power vs Speed curve of the Wind Pumping System for monthly average
wind speeds at (a)Kutubdia (b)Patenga (c)Sathkhira

4. Result Analysis
4.1 Discharge of the WPS
From the output power of the WPS, the discharge of the pump is calculated from
the following equation:
w gHQ
Po (4.1)

Where, Po is the output power of the WPS, w is the density of water =1000kg/m3,
H is the head of the pump, Q is the discharge and is the efficiency of the pump.
The calculated discharge of the pump is then plotted against monthly average wind
speed at the three locations.

Fig 4.1 Discharge of the WPS (m3/s) vs. wind speed curve for monthly average
wind speed at (a) Kutubdia (b) Patenga (c) Sathkhira
From Table 3.1, the lowest monthly average wind speed at Kutubdia, Patenga and
Sathkhira are 4.7m/s, 5.91m/s and 2.81 m/s respectively. So the cut in wind speed
of the turbine[9] is taken as 4m/s for Kutubdia and Patenga and 2 m/s for Sathkhira
as shown in Fig 4.1.
4.2 Overall Efficiency of the WPS
The maximum power delivered by the turbine is found from the peak points of the
turbines power vs. rotational speed curve. The output power of the pump is found
from the intercept points of the pump and turbine curves. The ratio of the pump
output power to the maximum turbine output power gives the overall efficiency, o
of the WPS.
output power of the pump
o ..(4.2)
maximum output power of the turbine
The performance curve of the WPS is obtained by plotting the product of the
overall efficiency and the maximum turbine power coefficient Cp against wind
speed v.

Fig 4.2 Overall performance curve of the WPS at (a) Kutubdia (b) Patenga (c)
Sathkhira
From the performance analysis of the system at three sites, it is found that the
system has a reasonably good efficiency at Kutubdia and Patenga for a long period
of the year. The overall efficiency of the wind pumping system at Sathkhira is
good for a very short time throughout the year. But it is also observed that the
maximum efficiency is obtained during the dry season of April and May, when
energy demand for irrigation purposes is high. So the wind pumping system for
Sathkhira is also lucrative.

5. Conclusion
The aim of the study was to design an optimum wind pumping system suitable for
harnessing wind energy at three coastal districts of Bangladesh- Kutubdia, Patenga
(Chittagong) and Sathkhira. A simulation process has been carried out to design a
wind turbine model using the QBlade software. A customized blade profile was
designed and the performance of the turbine at different wind speed regime was
simulated to obtain the characteristic curve of the turbine. A study of selecting the
suitable type of pump for the wind pumping system was conducted and a
centrifugal pump was chosen for the system. The output power of the WPS was
estimated plotting power against rotational speed of both turbine and pump in a
graph for the three selected sites. Furthermore, an overall performance evaluation
of the wind pumping system at the three sites was done. In this study, a simple
blade profile geometry was used for generating the turbine blade in order to reduce
cost. In our future work, we will use a more complex blade profile with varying
twist angle to ensure high performance and efficiency of the turbine at various
wind speeds.

6. References
[1] Muhammad Mahbubul Alam, Wind-Driven Rotodynamic Pumps For Low-
Lift Applications, Phd thesis, Department of Engineering, University of Reading,
UK
[2] M.J. Khan, M.T. Iqbal, S. Mahboob, A Wind Map of Bangladesh, Faculty of
Engineering, MUN, St. Johns, NL, Canada A1B3X5
[3] Robert Gasch, Rudiger Kortenkamp, Joehen Twele, A Simple Method for
Near Optimum Design of Wind Turbines With Centrifugal Pumps
[4] QBlade Guidelines, v0.6
[5] Hydraulic Machines, Dr Md. Quamrul Islam
[6] Dixon S.L. Fluid Mechanics, Thermodynamics and Turbo machinery , 5th
Edition
[7] Jhon McCosker, Design and Optimization of a Small Wind Turbine, Masters
of Engineering in Mechanical Engineering project, Rensselaer Polytechnic Institute,
Hartford, Connecticut, December 2012
[8] Plot Digitizer http://plotdigitizer.sourceforge.net/
[9] T. Al-Shemmeri, Wind Turbines