Energy 36 (2011) 2360e2366

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Modeling and optimization of the chamber of OWC system

Nader Dizadji a, *, Seyed Ehsan Sajadian b
a
Mechanical Engineering Department, Science and Research branch, Islamic Azad University, Tehran, Iran
b
Graduate of Energy Department, Science and Research branch, Islamic Azad University, Tehran, Iran

a r t i c l e i n f o a b s t r a c t

Article history: Due to their non-polluting nature and environment friendliness, Renewable Energies have gained great
Received 6 June 2010 deal of attention and deserve a substantial body of both theoretical and empirical research. Amongst
Received in revised form other factors, the low operational cost and simple maintenance procedures attributed the Oscillating
4 January 2011
Water Column (OWC) are perhaps the main reasons why this system is the most used concept for the
Accepted 9 January 2011
Available online 22 April 2011
ocean wave energy capture.
In this paper, through extensive experimental research various geometrical designs of an OWC system
is investigated and the optimized set up for the maximum energy harness is obtained.
Keywords:
Renewable Energy
The initial chamber dimensions were 10
50
53 cm with the chamber being placed in an open
Wave Energy channel with wave-simulating equipment with dimensions of 16
0.7
.05 m. For various chamber
OWC geometries, with the aid of a air rotameter and a Pitot tube equipped with a digital manometer, the outlet
air flow and velocity from the chamber was measured and registered.
The measurements were then interpreted to provide design data for the optimal geometry of the
chamber that may yield the maximum conversion of wave energy to useful energy.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Pursuant to the 1970’s oil crisis, attention was paid to the har-
nessing of energy from OWC’s and accordingly a 70 KW chamber
Renewable energies have gained more relevance and impor- was designed and erected under British university research fund-
tance in the energy producing sector. This is partly because they are ing, which later at year 2000 resulted in a 500 KW plant which was
classed as Clean Energy, as opposed to fossil fuel based energy located in shorelines of Western Scotland [3]. The simple features
production, and partly due to their immediate and long-lasting and design of this plant became the basis for many future economic
availability. In many nations around the world, policies are being OWC systems [4].
initiated to raise the share of renewable energy as fast as possible. The device consists essentially of a bottom-fixed structure,
In March 2007, the European Union defined a target of 20% whose upper part forms an air chamber of rectangular cross-
renewable energy for year 2020. In Denmark, a target of 30% section and whose immersed part being open to the action of sea
renewable energy for year 2025 has just been proposed by the water. The reciprocating flow of air displaced by the inside free
Danish Government [1]. There is a growing perception by society of surface motion ultimately drives a bi-directional turbine attached
the risks of dramatic global climate changes due to anthropogenic to the top of the structure. In such arrangement, as indicated in the
greenhouse gases, in particular energy related emissions of CO2. Fig. 1, a generator unit coupled to the turbine produces the intended
This has spurred a renewed interest in carbon-free or carbon- electrical energy [5].
neutral technologies for converting sources of renewable primary The power train in the so-described OWC consists of the wave
energy to electricity and to transportation fuels. However, it takes energy transfer to the oscillating air column in the chamber, fol-
energy to produce energy, even when the primary source is ener- lowed by the transfer of the energy embedded in the oscillating
getically cost-free, such as solar or wind[2]. motion to the mechanical rotation of the turbine shaft which itself
The oscillating wave column (OWC) wave energy device is drives a generator which converts the induced rotation into elec-
probably the most widely researched type of wave based power trical energy [3].
plants. The overall efficiency of an OWC depends on the individual
efficiencies of the oscillating air column and the turbine itself [6].
By installation of various bi-directional turbines, their influence on
* Corresponding author. Tel.: þ98 4486 9724; fax: þ98 4486 9734. the overall operation and efficiency of OWC systems was widely
E-mail address: [email protected] (N. Dizadji). compared and studied.

0360-5442/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.energy.2011.01.010
 N. Dizadji, S.E. Sajadian / Energy 36 (2011) 2360e2366 2361

Nomenclature

H wave height within the channel

Q outflow rate of air
l wave length
d water depth within the channel
ΔP pressure difference
V velocity of air outflow
r air density
D diameter of Outflow-air Tube
T wave time-Period

In the bi-directional turbines, the symmetrical placement of

turbine blades results a uni-directional turbine shaft rotation [7, 8].
Influence of blade arrangement, guide vanes, and non-return valves
on the turbine efficiency was subject of several studies [9]. Dissi-
pation of energy and power output of turbines for a fixed geometry
Fig. 2. Open channel in the lab.
of the chamber at various air flows was also greatly studied [10].
The main purpose of this experimental work was to design and
fabricate several chamber geometries which were then placed in an
open channel in which the wave motion was simulated artificially
in Science and Research branch, Islamic Azad University of Iran. The
outflow rate of air, contained in the chamber, was then measured
by accurate and correctly calibrated Pitot tube, and a rotameter. The
experimental results were then utilized to indicate a chamber
design and geometry which provided optimal energy output. At
larger device scales, it is expected that power plant will be
commercial. This study was set to be independent turbine type.

2. Theory

The important non-dimensional numbers in the designing of

our introduced OWC chamber are defined in this section [11].
The ratio of the cross-sections of air tube to that of oscillating
water column:
An
(1)
Ax
The ratio of the length of the chamber located in parallel to the
wave direction to the wave length: Fig. 3. Wave stimulator.

X+
(2)
l
The ratio of the incoming wave height to the wave length:

H
(3)
l
The Chamber Efficiency [12]:

Fig .1. OWC system [9]. Fig. 4. Modeled chamber in the channel.
2362 N. Dizadji, S.E. Sajadian / Energy 36 (2011) 2360e2366

power of output air Pu

h¼ ¼ (4)
power of wave energy Pw
where power of wave energy is derived from:

rw ¼ 1:225rgH2 (5)
where rw: water density.
Power of output air is calculated from the following
relationship:
 
1 pD2
Pu ¼ P þ rV 2 V (6)
2 4
where P: pressure of air at top of the channel, r: air density ,V: out
flowing air velocity, D: diameter of outflow air tube.
The chamber efficiency is thus related to the wave-height and
the outflow velocity of air.

Fig. 5. Map of chamber (dimensions are in centimeter).

3. Experimental setup and procedure

The experimental setup consisted of an open channel of

16
0.7
.05 m dimension (Fig. 2), equipped with a controllable Table 1
The results of the first series of the experiments (b ¼ 10cm).
paddle-driven wave simulator placed at one end of the channel
(Fig. 3). An artificial seashore preventing wave-return was situated Experiments Re ave Re max Qave(lit/s) Qmax(lit/s) DPave(pa) DPmax(pa) q(deg)
at the other end of the channel. 1 9319 16410 4.1 7.3 12.2 37 60
Making use of the dimensional analysis technique related Eqs. 2 10940 24312 4.9 10.8 16.3 82 50
3 9117 17828 4 7.9 11.3 43.5 40
(1)e(4), the utilized OWC was designed and fabricated. The OWC
4 15600 28971 6.9 12.9 33.4 114.4 30
was made from Plexiglas secured within an Aluminum frame. The

Fig. 6. Time-pressure diagrams (Pa-s).

 N. Dizadji, S.E. Sajadian / Energy 36 (2011) 2360e2366 2363

Table 2
The results of the second series of the experiments (b ¼ 10cm).

Experiments Re ave Re max Qave(lit/s) Qmax(lit/s) DPave(pa) DPmax(pa) q(deg)

5 11548 22488 5.1 10 18.2 69.2 60
6 10834 25122 4.9 11.2 16.3 86 50
7 14587 29984 6.5 13.4 29.6 122.2 40
8 13979 29579 6.2 13.2 26.7 120 30

manometer (both instruments were accurately calibrated in fluid

mechanics laboratory). Ultimately, making use of the registered
flow rate readings, and the actual diameter of the tube, the
Reynolds Number (Re) of the outlet air flow from the Chamber
was calculated by use of Eq. (7), and then the flow regime was
determined.

rVD
Re ¼ (7)
m
Fig. 7. Map of chamber (dimensions are in centimeter).
4. Result
structure was so designed and made that allowed for variation of
incident angle, and other critical dimensions .The design should be Since the chamber’s outflow of air was oscillatory, the respective
optimized for easy debugging, troubleshooting and repair (Fig. 4). measured pressure difference and air velocity were also oscillatory.
Appropriate sealing material was used to prevent air and water For each experimental run from the relation (7) Re was calculated,
leakage through the Aluminum frame. As indicated in Fig. 5, a tube the ΔP-t diagram was plotted, and the mean pressure difference and
was mounted at the top of the chamber to guide the exit chamber- mean velocity were then used for actual calculations.
air toward the intake of the turbine. In all calculations the following fixed values for the simulated
All experimental readings were taken at the steady state waves were considered:
conditions. Air flow is measured with an air rotameter and
velocity readings were registered by a Pitot tube with a digital H ¼ 6 cm; T ¼ 1:7 s; l ¼ 3:1 m; d ¼ 52 cm

Fig. 8. Time-pressure diagrams (Pa-s).

2364 N. Dizadji, S.E. Sajadian / Energy 36 (2011) 2360e2366

Table 3
The results of the 3rd series of experiments (b ¼ 20 cm).

Experiments Re ave Re max Qave(lit/s) Qmax(lit/s) DPave(pa) DPmax(pa) q(deg)

9 15195 29174 6.8 13 31.3 115.7 60
10 14992 29377 6.7 13.1 30.7 117.8 50
11 16208 30390 7.2 13.6 36.1 125.6 40
12 17018 32821 7.6 14.7 40 147 30

Six sets of experiments were conducted and the procedures and

their respective setups are grouped and explained in the following
six sections:

4.1. The first group of experiments

Referring to Fig. 5, for the first set of conducted experiments, the

back plate of the chamber was kept at a fixed 50 angle with respect to
the horizontal plane. The front-end plate was then placed stepwise at
angles q equal to 30 , 40 , 50 , and 60 with respect to the horizontal
plane. For every q, the chamber air’s outflow was measured and
recorded. Referring again to the same Figure (Fig. 5), the length "b" of
the topside of the chamber was set at 10 cm. For each set of experi-
ment, the ΔP-t diagram was then plotted, and with the aid of the same
diagram the mean and the maximum pressure difference were
Fig. 10. Map of chamber (dimensions are in centimeter).
calculated. The results obtained from the first series of the experi-
ments are shown in Table 1. Based on this table the air flow at q¼30 is plane, with angle q varied stepwise at 30 , 40 , 50 , and 60 with
maximum. Accordingly, Fig. 6 presents the ΔP-t diagram. respect to the horizontal plane. Then, the same procedure for
calculation of the mean and maximum pressure difference was
4.2. The second group of experiments followed. The results of the second series of the experiments are
shown in Table 2. Based on this table the maximum air flow rate is
Referring to Fig. 7, for the second set of experiments, the back gained at q¼40 .The ΔP-t diagram for this set of experiments is
plate of the chamber was kept at 90 with respect to the horizontal shown in Fig. 8.

Fig. 9. Time-pressure diagrams (Pa-s).

 N. Dizadji, S.E. Sajadian / Energy 36 (2011) 2360e2366 2365

Table 4 Table 5
The results of the 4th series of experiments (b ¼ 20cm). The results of the 5th series of the experiments (b ¼ 20 cm).

Experiment Re ave Re max Qave(lit/s) Qmax(lit/s) DPave(pa) DPmax(pa) q(deg) Experiment Re ave Re max Qave(lit/s) Qmax(lit/s) DPave(pa) DPmax(pa) q(deg)
13 17626 33429 7.8 14.9 42.8 152 30 14 7698 26338 3.4 11.8 28.1 94 30

Table 6
The results of the 6th series of the experiments (b ¼ 20 cm and H ¼ 6.8 cm).

Experiment Re ave Re max Qave(lit/s) Qmax(lit/s) DPave(pa) DPmax(pa) q(deg)

15 22691 39912 10.1 17.8 70 216 30

at 30 with respect to the horizontal plane, and the back plate is
held up right. Here b ¼ 20 cm. The results of the fourth series of
experiments (experiments 9 through 12), are shown in Table 4, and
the associated ΔP-t diagram is indicated in Fig. 11.

4.5. The fifth group of experiments

In all previous experiments the chamber’s air flow rate is

measured only during compression. In this series, the air flow is
measured in the expansion period and compared to the com-
pressing period. For this group of experiments the same set up
depicted in Fig. 10 is used. The results of the fifth series of experi-
ments (experiments 13 through 16) are shown in Table 5, with its
Fig. 11. Time-pressure diagram (Pa-s).
resulting ΔP-t diagram shown in Fig. 12.
4.3. The third group of experiments
4.6. The sixth group of experiments
In the third series of the experiments the length "b" of the top
side is doubled (b ¼ 20 cm), with the back plate of the chamber In the final series of experiments, the effects of variation of the
maintained fixed and at an angle of 90 with respect to the hori- properties of the incoming wave and air flow during the expansion
zontal plane, and the angle of the front plate with respect to the period is measured and compared. The same set up as indicated in
horizontal plane varied in steps from 30 , 40 , 50 to 60 . Then, the Fig. 10 is used.
same procedure for calculation of the mean and maximum pressure Initially, only the height of the incoming wave was varied, and
difference was followed. Table 3 presents the results obtained from then its length was also changed. At first, the height of the incoming
conducting the third series of experiments. The respective ΔP-t wave was increased by 10%. Here the properties of the incoming
diagram is shown in Fig. 9. wave were:

H ¼ 6:8 cm T ¼ 1:4s l ¼ 2:5 m d ¼ 52 cm

4.4. The forth group of experiments
Table 6 and Fig. 13 indicate the results attributed to these
Fig. 10 indicates the set up for the forth series of the experi- experiments for the set increased height of the incoming water
ments. In these experiments the front plate of the chamber is fixed waves.

Fig. 12. Time-pressure diagram (Pa-s). Fig. 13. Time-pressure diagram (Pa-s).
2366 N. Dizadji, S.E. Sajadian / Energy 36 (2011) 2360e2366

set of experiments. In the third series of experiments, the

maximum air flow rate occurred at is at q¼30 , in comparison to
the first and second set of experiments, the overall outflow of
chamber air was increased.
With regard to the fourth series it can be observed that posi-
tioning the outflow air tube on the top side of the chamber
increases the air outflow rate (refer to the 12th experiment).
Accordingly, one may deduce that mounting the turbine vertically
at the top of the chamber increases the power output; however,
placement of the turbine must be in a manner to avoid air-water
mixing leading to the presence of water droplets which can damage
the turbine blades.
According to the result of the fifth series it is seen that the air
outflow rate in the expansion phase is decreased by 20% as compared
with the previous set of experiments. This can be contributed to the
so-changed amplitude’s of the incoming water waves.
In the final series of experiments (experiments 17 through 20),
the height and length of the incoming water waves are varied. Here
the outflow of air is increased by 16% after increasing height of
wave as compared to the same condition prevailed at the experi-
Fig. 14. Time-pressure diagram (Pa-s).
ment Number 13.
It can be observed after increasing length of wave the outflow of
air is increased by 11% as compared to the similar conditions
Table 7 maintained during experiment Number 13.
The results of the 6th series of the experiments. (b ¼ 20 cm) (l¼2.9 m). From the results, one may deduce that the OWC’s working
Experiment Re ave Re max Qave(lit/s) Qmax(lit/s) DPave(pa) DPmax(pa) q(deg) efficiency is highly dependent upon the chamber’s geometry, and
16 19854 39101 8.8 17.5 53.9 208 30 the properties of the incoming water wave. It can be further
deduced that the chamber geometry can be systematically
changed to augment the outflow of air and thus the overall effi-
Then the wave length of the incoming wave was increased by ciency of the given OWC. The results are independent of turbine
30%. Here the properties of the incoming wave were: type and may be extended to similar case studies. The best
chamber’s geometry related to experiment number 13.In this case
H ¼ 6:1 cm T ¼ 1:5 s l ¼ 2:9 m d ¼ 52 cm efficiency of chamber corresponding Eqs. (4) to (6) was calculated
32%.It could be considered the following measurement maximum
Table 7 and Fig. 14 indicate the results attributed to the set of
uncertainty 1% ,1%,4% in pressure difference, velocity and
experiments in which the wave length of the incoming water waves
flow rate respectively.
is increased.

5. Conclusion References

[1] Lund H, Mathiesen BV. Energy system analysis of 100% renewable energy
Making reference to the results of the first set of experiments systems, the case of Denmark in years 2030 and 2050. Energy
(experiments 1 through 4), decreasing the front-end angle q 2009;34(5):524e31.
[2] Gonçalves da Silva C. The fossil energy/climate change crunch: can we pin our
considerably increases the outflow of air from the chamber. It
hopes on new energy technologies. Energy 2010;35(3):1312e6.
was also observed that the parallel arrangement of the front-end [3] Paixao Conde JM, Gato LMC. Numerical study of the air-flow in an oscillating
and back-end plates resulted in a relative maximization of the water column wave energy converter. Renewable Energy; 2008; doi:10.1016/
outflow of the chamber air. Although, in the first set of experi- j.renene.2008.02.028.
[4] Zhu SP, Mitchell L. Diffraction of ocean waves around a hollow cylindrical shell
ments the maximum air outflow resulted from the 4th experi- structure. Wave Motion 2009;46:78e88.
ment, the simulated wave pattern, in particular the respective [5] Jayashankar V, Anand S, Geetha T. A twin unidirectional impulse turbine
difference in wave height, and limitations imposed by the phys- topology for OWC based wave energy plants. Renewable Energy
2009;34:692e8.
ical built of the fabricated OWC system, which was itself attrib- [6] Setoguchi T, Takao M. Current status of self rectifying air turbines for wave
uted to the budget cut-backs of the project itself proved to be an energy conversion”. Energy Conversion & Management 2006;47:2382e96.
obstacle to the scale that only limited study on the parallel [7] Thakker A, Abdulhadi R. The performance of wells turbine under bi-direc-
tional airflow. Renewable Energy; 2008; doi:10.1016/j.renene.2008.02.013.
arrangement of the front and back-end plates were justified and [8] Thakker A, Hourigan F. Computation fluid dynamics analysis of a 0.6m,
performed. 0.6hub-to-tip ratio impulse turbine with fixed guide vanes. Renewable Energy
With regards to the second set of experiments (experiments 5 2005;30:1387e99.
[9] Brito-Melo A, Gato LMC, Sarmento AJNA. Analysis of wells turbine design
through 8), where the back plate is at 90 relative to the horizontal parameters by numerical simulation of OWC performance. Ocean Engineering
plate, in comparison to the first set of experiments, the overall 2002;29:1463e77.
outflow of chamber air was considerably increased. The same [10] Marjani AEL, Castro Ruiz F, Rodriguez MA, Parra Santos MT. Numerical
modeling in wave energy conversion systems. Energy 2008;33:1246e53.
pattern of increased outflow of chamber air with decreasing the
[11] Arlitt RGH, Tease K, Starzmann R, Lees J, 2007. “Dynamic system modeling of
angle q was again observed. It was notable that the same financial an oscillating water column wave power plant based on characteristic curves
constraints imposed limitations on the extent to which the study obtained by computational fluid dynamics to enhance engineered reliability”.
was conducted and as a result a more comprehensive research on Proceeding of the 7th European wave and tidal energy conference, Porto,
Portugal.
the so-designed and the so-fabricated OWC system deserves more [12] Falnnes J. Ocean waves and oscillating systems. Cambridge University Press,
attention. The associated costs were relatively higher for the second ISBN 0 521 78211 2; 2002.