Passive cooling technology for photovoltaic

panels for domestic houses

..............................................................................................................................................................

Shenyi Wu* and Chenguang Xiong

Department of Architecture and Built Environment, University of Nottingham, University
Park, Nottingham NG7 2RD, UK
.............................................................................................................................................
Abstract
The efficiency of photovoltaic panels decreases as the panels’ temperature increases, which results in
deduction of electricity generation. In order to reduce this effect, different cooling methods were

Downloaded from https://academic.oup.com/ijlct/article-abstract/9/2/118/633328 by guest on 20 July 2020

proposed and investigated. This paper reviews the previous work on cooling PV cells and concludes that
the cost-effectiveness, design feasibility and minimal energy consumption are the important design
consideration for cooling systems. Based on these considerations, this paper reports a passive cooling
method that utilizes rainwater as cooling media and a gas expansion device to distribute the rainwater.
The gas is thermally expanded from receiving solar radiation as such the amount of water it pushes to flow
over the PV cells is proportional with the solar radiation it received. The paper reports a design and
simulation of such a system for a domestic house application. In the paper, a relationship of the gas
chamber size, solar radiation and gas expansion volume was established for evaluation with respect to the
variation of gas temperature and the amount of rainwater used for cooling. A heat transfer model was
used to evaluate the performance of the cells by cooling with this passive device. The results show that on
a design day, the passive cooling system reduces the temperature of the cells and increases electrical
efficiency of the PV panel by 8.3%. The payback period of this system is ,14 years.

*Corresponding author. Keywords: PV cooling; passive cooling; solar pump; rainwater harvesting; heat transfer
[email protected].
uk Received 13 November 2013; revised 13 January 2014; accepted 10 February 2014
................................................................................................................................................................................

1 INTRODUCTION single cells passively. Araki et al. [4] did a further research on
passive cooling technologies and found that good thermal con-
The operating temperature is one of the important factors that duction between cells and heat spreading plate was important.
can affect the efficiency of the PV panels. The effects of tempera- Combining PV and solar thermal collectors (PV/T) is another
ture on photovoltaic efficiency can attribute to the influences on way of cooling PV panels. Tonui and Tripanagnostopoulos [5]
the current and voltage of the PV panels. This can be easily reported their experiment on modified PV/T collectors, and
found on the I-V curve of the panels. It results in a linear reduc- results showed the maximum temperature reduction achieves
tion in the efficiency of power generation as temperature 108C by natural ventilation and 308C by forced ventilation.
increases [1]. The efficiency of some types of PV cells is very As a good cooling media, water has been widely used for PV
much dependent on their operating temperature. For crystalline cooling in various forms. It is very suitable for PV/T systems.
silicon solar cells, the reduction in conversion efficiency is Kalogirou [6] studied a water-based PV/T system consisting of
0.4 – 0.5% for every degree of temperature rise [2]. Therefore, four monocrystalline PV panels in the Cyprus and achieved an
reducing the operating temperature of photovoltaic cells is increase of average annual electrical efficiency from 2.8 to 7.7%
important for the PV panel to work efficiently and protect cells with the payback periods of 4.6 years. Tripanagnostopoulos et al.
from irreversible damage. [7] compared electrical efficiency of PV/WATER, PV/AIR and
A number of researchers have worked on cooling the PV PV/FREE and PV/INSUL under ambient air temperature of
panels with different approaches. Air circulation is probably the 298C. They achieved the maximum increase by 3.2% with PV/
most simple and natural way for this purpose. In order to WATER.
enhance convection heat transfer, fins were used to extend the Krauter [8] investigated the method of covering PV modules
heat transfer area. Edenburn [3] developed a device, made up of with a flowing water film above. With the additional evaporation
linear fins on all available heat sink surfaces, used for cooling heat transfer, it was claimed that they could decrease the cell

International Journal of Low-Carbon Technologies 2014, 9, 118–126

# The Author 2014. Published by Oxford University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which
permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
doi:10.1093/ijlct/ctu013 Advance Access Publication 26 March 2014 118
 Passive cooling technology

temperature up to 228C and obtained a net increase from 8 to

9%. Abdolzadeh and Ameri [9] used water spray to cool the PV
panels and achieved increasing the efficiency of cells by 3.26 to
12.5%. Kordzadeh [10] studied that a thin continuous film of
water running on the front of the surface of modules obtained
better electrical efficiency because of reducing reflection loss and
surface temperature.
To avoid additional energy consumption incurred for cooling
the PV panels, Furushima and Nawata [11] reported a model
with cooling water being supplied from a city water supply
system by Siphonage and the cooling system did not require any
additional energy input on the site. Wilson [12] studied the
gravity-fed technology where water was transported from up-

Downloaded from https://academic.oup.com/ijlct/article-abstract/9/2/118/633328 by guest on 20 July 2020

stream sources like river to downstream sources by gravity. The
results obtained from this work showed a 12.8% increase in elec-
trical efficiency as a result of 328C temperature reduction.
Other technologies were also used to enhance the heat trans-
fer for cooling the panels. Akbarzadeh and Wadowski [13]
reported an innovative gravity-assisted heat pipe system to opti- Figure 1. 3-D model of the solar-driven rainwater cooling system installed on
mize the cooling of concentrated photovoltaics. It was found the roof.
that the temperature at the surface of solar cells did not exceed
468C during a 4-h test, and the efficiency was increased by 50%.
Huang et al. [14] initially integrated PCM into BIPV system and
used fins for improvement. Biwole et al. [15] established a nu- south- and north-facing roofs in order to maximize the rain-
merical model and used CFD to simulate heat and mass transfer water harvesting. On receiving the solar radiation, the gas in the
of PCM at the back of photovoltaic panels. Their results showed chamber expands with the temperature increase. The rainwater
that adding PCM at the back of panels can maintain the operat- in the tank is pushed upwards by the expanding gas so that it
ing temperature below 408C. flows over the PV panel through a distribution tube on the top
Active cooling is effective to cool PV panels. However, with as shown in Figure 2. The rainwater is not considered being
the additional power consumption involved, the active cooling reused to reduce the cost and simplify the system structure in
purely used to lower the operating temperature does not have this case.
obvious benefit in the net gain of efficiency. The technologies
such as PV/T ( photovoltaic thermal) system or the PV-SAHP
( photovoltaic solar heat pump) system [16, 17] seem to address 3 ANALYSIS OF THE SYSTEM
the issue stated earlier by combination of two systems. But the
fact that PV/T has to at a higher operating temperature in order The amount of the rainwater delivered to the PV panels is deter-
to supply useful heat means the gain by cooling is limited. What mined by the gas expansion volume. The expansion volume is a
is more, the higher initial investment and the final benefit with function of temperature that varies with the solar energy the gas
PV/T technology is contributed to thermal energy rather than received. Therefore, there is a relationship between the amount
electricity [7]. This renders the PV/T being not an effective tech- of rainwater delivered and solar incidence. The relationship can
nology for the original purpose. Therefore, finding a simple and be derived from the energy conservation law.
feasible way to cool the PV panel without requiring further
energy input is still much sought after.
3.1 Energy balance equation in the gas expansion
chamber
2 DESCRIPTION OF THE SYSTEM The gas expansion chamber is covered by an insulation layer
used to reduce heat loss from the side surfaces and is covered with
Figure 1 illustrates an example of the proposed solar-driven rain- an absorption layer to enhance the capture of solar radiation.
water cooling system. The system consists of a PV module with Taking the chamber as a control volume, the energy conversion
an area of 1.46 m2, maximum efficiency of 15.4% and maximum and heat flows of the chamber are as shown in Figure 3. The
power output of 250 W, a gas expansion chamber, a rainwater energy balance can be expressed as Qnet ¼ G  Qc  Qr  WE ;
storage tank and a secondary water tank. A cylindrical gas ex- where WE is gas expansion work (J). The Qnet is the heat that
pansion chamber is installed at the eaves whereas the secondary causes the gas temperature rise.
water tank, which is connected to the gas expansion chamber, is To simplify the simulation, some assumptions are made and
hung at the side of the house. Gutters are installed on both stated as follows:

International Journal of Low-Carbon Technologies 2014, 9, 118– 126 119

S. Wu and C. Xiong

Downloaded from https://academic.oup.com/ijlct/article-abstract/9/2/118/633328 by guest on 20 July 2020

Figure 2. Schematic diagram of cross section of the solar-driven rainwater cooling system.

K); a is absorption coefficient of PV panel (0.95); 1 is emissivity

factor of absorber (0.04); s is Stefan – Boltzmann constant
(W/m2 K4); t represents time (s).
This equation describes the accumulating effect of the solar
radiation on the gas expansion chamber from a reference point
and its derivative form:

4 dV
Aab aI  Aab hc ðTgas  Ta Þ  Aab 1sðTgas  Ta4 Þ  Patm
dt
Figure 3. Schematic diagram of cross section of the gas chamber. ð2Þ
dT
 mgas Cgas ¼0
dt
† No heat transfer across the side boundaries. The conduction
heat transfer between the absorption layer and the surround- It describes the effect of the solar radiation on the gas chamber
ing air is neglected. at any time point. Since the chamber’s temperature change is a
† Thermal resistances in the absorption layer and gas are not slow process, we use the finite-difference equation to approxi-
considered so that the absorption layer and gas have the same mate Equation (2) as follows:
temperature and the gas temperature in the chamber is
uniform. 4
Aab aI  Aab hc ðTgas  Ta Þ  Aab 1sðTgas  Ta4 Þ
† The gas is treated as ideal gas.
† The atmospheric pressure is constant within duration of time. Patm ðAw HðtÞ  Aw Hðt1Þ Þ

† The gas properties are constant. Dt
mgas Cgas ðTgas;ðtÞ  Tgas;ðt1Þ Þ
Under the assumptions, the energy balance of the gas expansion ¼ ð3Þ
Dt
chamber can be expressed in the following equation:
ð ð where the subscribes (t) and (t 2 1) denote the time step in
Aab a I dt  Aab hc ðTgas  Ta Þ dt hour and AW is the base area of the secondary tank (m2) and H
ð is the height of water pumped (Figure 4).
4
 Aab 1s ðTgas  Ta4 Þ dt  Patm dV For 1-h time interval, we have the following equation:

¼ mgas Cgas DTgas ð1Þ 4

Aab aI  Aab hc ðTgas  Ta Þ  Aab 1sðTgas  Ta4 Þ
Patm ðAw HðtÞ  Aw Hðt1Þ Þ
where I is solar radiation on horizontal surface (W/m2); Tgas 
represents gas temperature (8C); Aab represents area of absorber 3600
surface (m2); Cgas denotes specific heat capacity of gas (J/kg K); mgas Cgas ðTgas;ðtÞ  Tgas;ðt1Þ Þ
¼ ð4Þ
hc denotes convection heat transfer coefficient of wind (W/m2 3600

120 International Journal of Low-Carbon Technologies 2014, 9, 118– 126

 Passive cooling technology

Downloaded from https://academic.oup.com/ijlct/article-abstract/9/2/118/633328 by guest on 20 July 2020

Figure 4. Schematic diagram of cross section of the gas chamber and secondary
water tank.

3.2 Hydraulic head of water in the secondary water

Figure 5. Schematic diagram of cross section of the PV panel.
tank
With reference to Figure 4, the expression of hydraulic head in
the secondary water tank varying with the gas expansion can be where G denotes energy generated from solar radiation (J),
derived as follows. U indicates internal energy (J), and Qe, Qc, and Qr denote heat
At initial state, the secondary water tank is filled with water loss by evaporation (J), heat loss by convection (J) and heat loss
and the gas in the chamber does not expand. The state of the gas by radiation (J), respectively.
can be expressed as follows: The heat transfer mechanism is quite complex due to
temperature variation along the water– solid interface. A two-
Patm Vgas ¼ mgas RT0 ð5Þ dimensional steady-state model is used, and some assumptions
should be made to simplify the calculation.
where T0 denotes initial temperature of the chamber (8C); Patm
indicates atmosphere pressure (Pa). † Solar radiation irradiates on the PV panel, 15.4% is converted
On receiving heat from the solar radiation, the gas in the to electricity energy, 5% is reflected and the rest part is con-
chamber starts to expand. If the gas volume is expanded by dV verted to heat energy.
(volume expansion) (m3), the same volume of water will be † Assume the water mass is uniformly distributed over the PV
pushed out of the tank. The change of the state follows the fol- panel and water is ultimately heated to a temperature that is
lowing equation: same as the cell temperature Tc.
½Patm þ rgð0:9 þ HÞðVgas þ dV Þ ¼ mgas RTgas ð6Þ † Water temperature increase caused by solar radiation is
neglected. [18]
where mgas represents mass of gas (kg); Vgas denotes volume of † Convective heat loss at the back of PV panel and radiation
gas (m3) and dV ¼ H  AW ; where AW ¼ 0:25: From Equations heat transfer is not considered.
(5) and (6), a relationship between the volume of the water
pushed and gas temperature can be obtained: Air flowing at the air– water interface essentially accelerates
water evaporation rate. Thus, the air convection could be mea-
rgAw H 2 þ ðPatm Aw þ rgVgas þ 0:9rgAw ÞH þ Patm Vgas sured accompanying with water evaporation. Smith et al. [19]
ð7Þ predicted evaporation heat transfer flux by the following equa-
Patm Vgas Tgas tion, which approximately estimates how much latent heat is
þ 0:9rgVgas  ¼0
T0 removed from the PV panel by water evaporation (w/m2).

3.3 Heat transfer on the PV panel qe ¼ ð0:0638 þ 0:0669VÞðPw  Pa Þ ð8Þ

When the water flows over a tilted PV panel, the heat transfer
between the water and panels can be complicated by involving where saturation pressure of water is as follows:
water evaporation in addition to the normal radiation and con-
vection heat transfer. With reference to Figure 5, the energy log ðPs /1000Þ ¼ 30:59051  8:2 logðTa þ 273:16Þ þ 0:00248ðTa
balance leads to:
3142:31
þ 273Þ 
G ¼ Qe þ Qc þ Qr þ DU ðTa þ 273Þ

International Journal of Low-Carbon Technologies 2014, 9, 118– 126 121

S. Wu and C. Xiong

partial pressure of water vapour at surrounding air is as follows:

Pa ¼ wPs

partial pressure of water vapour at water is as follows:

 
5132
Pw ¼ exp 20:386   133
0:5ðTw;in þ Tc Þ

where Tw,in denotes water inlet temperature (8C). Figure 6. Accumulated rainwater volume pushed by gas chambers with
Therefore, total heat transfer via water evaporation can be different sizes.
expressed as follows:

Qe ¼ Acell qe ð9Þ area. But the increase is within 18C. This could be a result of the

Downloaded from https://academic.oup.com/ijlct/article-abstract/9/2/118/633328 by guest on 20 July 2020

heat gain from the larger surface area being offset by the heat
where Acell denotes area of cells (m2). losses from the same larger surface area. Figure 6 presents accu-
The internal energy change of water can be expressed as mulated water volume pushed due to gas expansion. It can be
follows: seen that the amount of water pumped increases with the gas
mwater Cp ðTw;out  Tw;in Þ mwater Cp ðTc  Tw;in Þ chamber volume. It increases from 123 l/day with a 0.75-m3
DU ¼ ¼ ð10Þ chamber to 200 l/day with a 1.5-m3 gas chamber. The amount
3600 3600
pumped increasing with the chamber volume is due to the as-
where Tw,out denotes water outlet temperature (8C). sumption of uniform air temperature in the chamber. The
Thus, after a water film flowing down to a PV panel, the cell volume expands more with the air mass increases in the
temperature can be approximately calculated as follows: chamber. However, this phenomenon should become less sig-
nificant when the air temperature profile in the chamber is
aIAcell ð1  hÞ ¼ Acell ð0:0638 þ 0:0669Uw ÞðPw  Pa Þ treated as non-uniform. Without any control, the gas chamber
mwater Ca ðTc  Tw;in Þ can pump the maximum amount of water to the PV panel at 7
þ ð11Þ am, and the amount gradually decreases to zero around 1 pm. It
3600
was estimated approximately that 165 l/day of rainwater is avail-
where Uw denotes wind speed (m/s). able for the climate under the consideration. To pump this
amount of water, a 1-m3 gas chamber is needed.

4 SIMULATION RESULTS 4.2 Performance of the PV module cooled by the

This system was analysed with the climate data on a clear day of
system
The performance is evaluated from the rainwater pumped by the
29th July in Nottingham [20]. The day was chosen for its low
device with a 0.16-m3 secondary water tank connecting to the a
wind speed and high air temperature. The study was focused on
1-m3 gas chamber. The secondary water tank is designed to
the thermal performance of the gas chamber and the cooling
protect gas infiltration and increase air tightness so that the
effect to the PV panels. The analysis was on one PV module
whole system can work with the higher efficiency. The variation
system.
of the gas temperature in the chamber during a day was illustrated
in Figure 7. The initial gas temperature is 293 K, which is
4.1 Influence of design parameters on the system quickly heated to 342 K at 7 am, because low heat capacity of
performance gas, small mass of gas in the chamber and intensive solar radi-
The size of the gas chamber is a predominant parameter that ation on the design day cause that gas can be heated to a high
influences how much water can be pushed out of a tank. To temperature. With the increase of solar radiation, the gas tem-
evaluate the influences of chamber surface area and chamber perature gradually increases to a maximum value 387 K at 1 pm.
volume on performance, two groups of the gas chambers were After that, gas temperature reduces due to the reduction of solar
analysed. In the first group, the chambers have the same surface radiation and the rise of heat loss from a gas chamber to the
area of 2.5 m2 but with different volumes of 1.5, 1.25, 1 and outside environment.
0.75 m3, respectively. The second group has a fixed volume of Under this condition, the device is able to push 152 l of water
1.25 m3 but with the different surface areas of 3, 2.5, 2 and to the PV panel (Figure 8). As discussed earlier, without control,
1.5 m2, respectively. In this system, water is pushed up through a the majority amount of the water is pumped at the early time of
0.9-m-high vertical pipe (Figure 4). the day when the demanding for cooling is low. A control to the
The results show that, for the same chamber volume, the gas flow may be needed, for example, by a temperature sensitive
temperature in the chamber slightly increases with the surface valve to delay the water pumping to address this issue. The

122 International Journal of Low-Carbon Technologies 2014, 9, 118– 126

 Passive cooling technology

Table 1. Climate data of the day of calculation.

Time Temperature Wind Humidity Beam solar Diffuse solar
(8C) speed (%) radiation radiation
(km/h) (w/m2) (w/m2)

7 16.3 6 80 517 122

8 18.2 4.6 72 528 207
9 20.3 4.3 67 634 193
10 22.3 4.3 62 744 166
11 24 4.3 59 712 232
12 25.2 6.5 56 682 249
13 26.2 6.5 54 798 149
14 27 5.4 52 768 153
15 27 4.3 51 558 216
16 26.7 7 53 646 128

Downloaded from https://academic.oup.com/ijlct/article-abstract/9/2/118/633328 by guest on 20 July 2020

17 25.6 10.3 57 381 134
18 24.2 6.5 61 463 59

Weather data on 29th July.

Figure 7. Gas temperature variation in a design day.

Table 2. Hourly rainwater supply with temperature sensitive control for
cooling.
Time Water volume (l)

10 28.37
11 30.13
12 30.62
13 31.57
14 31.31

Figure 8. Accumulated rainwater volume in each hour.

operating temperature of PV is primarily determined by the

solar radiation. On the day of 29th July, between 10 am and 14
pm, solar radiation was .850 W/m2 and its temperature
reached 508C. The detail of the climate data for the day is shown
in Table 1. In order to maximize the cooling benefit to the PV
panel, a temperature sensitive valve can be used to adjust the
flow rate of water according to the roof temperature. Table 2 Figure 9. Comparison of operating temperatures between with cooling and
shows that a total of 152 l of water can be pushed at different without cooling to the PV panel.
hourly rates with respect to the roof temperature from 10 am to
2 pm on the day. It can be seen that with the temperature sensi- without cooling, respectively. The cooling maintains the efficiency
tive valve, more water is pumped when the roof temperature is of the cells above 14.5% each hour in a design day, particularly,
higher at late hours, which allows more cooling to the PV penal between 12 pm and 2 pm during which the PV panel has very low
when it receives high solar radiation. efficiency without cooling. The cooling also increases the power
During the working time, the cooling to the PV panel is very output by 16 W on average. In summary, solar-driven cooling
effective when the PV panel temperature is high as shown in system is able to reduce the operating temperature of the cells by
Figure 9. It can be seen that at 1 pm, a maximum temperature re- 16.58C on average, and it has a better cooling effect when the tem-
duction of 198C is achieved and at other time temperature perature of the cells becomes higher. In addition, daily electrical
reduction ranges from 12.5 to 18.58C. Figures 10 and 11 present yields of the PV module will grow 80 Wh, achieving an increment
the efficiency and the power output of the PV panel with and of 8.3%. However, variable environmental conditioning has

International Journal of Low-Carbon Technologies 2014, 9, 118– 126 123

S. Wu and C. Xiong

Figure 12. Volume of daily pushed rainwater by a gas chamber in different

months.

Downloaded from https://academic.oup.com/ijlct/article-abstract/9/2/118/633328 by guest on 20 July 2020

Figure 10. Comparison of the efficiencies between with cooling and without
Table 3. Comparison between collected rainwater and required rainwater.
cooling to the PV panel. Month Daylight Equivalent Operating Rainwater Required
hours sunny day day ratio collected rainwater
(l/day) (l/day)

April 146.32 14.60 0.95 123 89

May 200.9 18.26 1.56 110 104
June 205.1 18.65 1.64 239 111
July 174.96 14.58 0.95 177 107
August 164.24 14.93 1 158 103
September 141.42 14.14 0.89 104 81

It is not efficient and cost-effective to design this solar-driven

rainwater cooling device to work every day, especially for the
rainy and cloudy days. Thus, equivalent sunny days in each
month can be predicted based on an assumption of 10 –12 sun
hours in a sunny day in different months. In an ideal scenario,
Figure 11. Comparison of the power outputs between with cooling and without sunny days and rainy days occur intermittently and an operating
cooling to PV panel. day ratio (number of sunny days/number of rainy days) is calcu-
lated to evaluate the relationship between collected rainwater
and required rainwater. Table 3 shows that, except in May, the
impacts on gas chamber expansion, so does on water pumping amount of collected rainwater can meet the requirement of the
and the cooling effect. Therefore, it is meaningful to evaluate cooling system in each month. A 1000-l water tank allows it to
annual performance of the solar-driven water cooling system meet the water consumption up to 10 days under the worst-case
under a stable environmental condition. scenario like continuous sunny days.

4.3 Monthly water consumption by the system

4.4 Annual energy saving and payback period
The supply of the rainwater depends on the gas expansion in a
The certain amount of rain water can cool more PV modules if
chamber, which varies with solar radiation and ambient air tem-
heat is removed by evaporation, under the premise that rain-
perature. As shown in Figure 12, in January and December, little
water uniformly covers the PV modules. Based on available
water can be pumped by this device; however, in June, intensive
rainwater in each month, it is estimated that this solar-driven
solar radiations and high air temperatures make the device to
rainwater cooling system can increase 33.4 kWh of electrical
pump 110.8 l of water to PV panel for cooling in each day.
yields for a domestic house when six PV modules are applied.
According to solar radiation and rainwater supply, the system
To comprehensively analyse the benefits of a new system, the
was designed to work between April and September.
economic analysis of the solar-driven water cooling system is
For a well-constructed roof, the runoff coefficient is usually
conducted by the extra cost of equipment required to con-
assumed as 0.8 [21]. Therefore, monthly rainwater collection
struct this cooling system, against additional energy benefits
can be estimated from the following equation:
obtained from the modified PV panels. The total cost of this
Rainwater volume ¼ monthly rainfall  catchment area passive cooling system is estimated as £197., i.e. £80 to the cost
of the rainwater harvest system and £117 to the cost of the gas
 runoff coefficient expansion chamber, a secondary water tank, pipes and valves.

124 International Journal of Low-Carbon Technologies 2014, 9, 118– 126

 Passive cooling technology

Table 4. Payback period analysis. 5 CONCLUSION

Year Power Electricity Feed-in-tariff Annual Total Initial
generated supply (£) at 2.5% operating saving capital This paper reports a passive cooling system, which can be used
at 1% rate (£) at inflation rate saving (£) (£) growth for cooling the PV modules on the roof of a domestic house in
discount 2.8% (£) at order to increase electrical efficiency. The simulation results for
rate inflation 3% this cooling system show:
(kWh) rate annual
interest
rate † The influences of the absorbing surface area on the water
supply volume are not obvious, whereas a gas chamber with
1 33.40 0.15 0.45 19.87 197.00 larger volume significantly increases the water supply.
2 32.73 0.15 0.46 19.98 202.91
3 32.40 0.15 0.47 20.28 209.00
However, the actual chamber size should be comprehensively
4 32.06 0.16 0.48 20.59 215.27 considered with roof area and available rainwater capacity.
5 31.73 0.16 0.50 20.90 221.73 † On the design day, the solar-driven rainwater cooling system

Downloaded from https://academic.oup.com/ijlct/article-abstract/9/2/118/633328 by guest on 20 July 2020

6 31.40 0.17 0.51 21.21 228.38 is able to pump 152 l of water to PV modules. The maximum
7 31.06 0.17 0.52 21.53 235.23 reduction in the temperature of the cells reaches 198C and
8 30.73 0.18 0.53 21.84 242.29
9 30.39 0.18 0.55 22.16 249.55
average electrical yield is increased by 8.3%.
10 30.06 0.19 0.56 22.48 257.04 † For the solar-driven rainwater cooling system operating
11 29.73 0.19 0.58 22.80 264.75 between April and September, this cooling system can in-
12 29.39 0.20 0.59 23.13 272.69 crease the electricity generation by 33.4 kWh annually.
13 29.06 0.20 0.61 23.45 280.23 280.87 † The simple payback period of the solar-driven rainwater
14 28.72 0.21 0.62 23.78 304.01 289.30
15 28.39 0.21 0.64 24.11 297.98
cooling system was found to be equal to 14 years under a con-
16 28.06 0.22 0.65 24.44 306.92 servative assumption. It still has potential and the initial cost
17 27.72 0.23 0.67 24.77 316.13 will be reduced if it incorporates with the guttering system.
18 27.39 0.23 0.68 25.10 325.61
19 27.05 0.24 0.70 25.44 335.38 The most significant point of this approach is that it utilizes
20 26.72 0.25 0.72 25.77 345.44
rainwater and solar energy to cool the PV panels—improving
21 26.39 0.25 0.74 26.10 355.80
22 26.05 0.26 0.76 26.44 366.48 PV system efficiency with no requirement for additional energy
23 25.72 0.27 0.77 26.77 377.47 input. The authors believe that it has the potential for further
24 25.38 0.27 0.79 27.10 388.80 exploration.
25 25.05 0.28 0.81 27.44 587.50 400.46

The saving in electrical yields per year equals to £20 when REFERENCES
feed-in tariff equals 0.45£/kWh and electricity rate equals
0.145£/kWh [22]. [1] Skoplaki E, Palyvos JA. On the temperature dependent of photovoltaic
A simple payback formula was used to calculate the payback module electrical performance: a review of effective/power correlations. Sol
period as follows with an inflation rate of 2.8% being taken into Energy 2009;83:614 –24.
account: [2] Notton G, Cristofari C, Mattei M, et al. Modeling of a double-glass photo-
voltaic module using finite differences. Appl Thermal Eng 2005;25:2854 – 77.
initial cost [3] Edenburn MW. Active and passive cooling for concentrating photovoltaic
Payback period ¼
annual operating saving arrays. Conference record, 14th IEEE PVSC, 1980, Washington State
Convention Center, USA, pp. 776– 776.
The annual saving in the equation is calculated from: [4] Araki K, Uozumi H, Yamaguchi M. A simple passive cooling structure and
Annual operating} saving = kWh  (electricity rate þ feed in its heat analysis for 500* concentrator PV module. Conference record, 29th
tariff ). Assume all the costs of this solar-driven rainwater system IEEE PVSC, 2002, Washington State Convention Center, USA, pp. 1568 – 71.
are paid up front; the power output of PV discount rate at 1% a [5] Tonui JK, Tripanagnostopoulos Y. Improved PV/T solar collectors with heat
year; the electricity inflation rate at 2.8%; the feed-in-tariff infla- extraction by forced or natural air circulation. Renew Energy 2007;32:
tion rate at 2.5% and annual saving rate at 3%. Based on the 623 – 37.
assumptions mentioned earlier, the calculation results are shown [6] Kalogirou AS. Use of TRNSYS for modelling and simulation of a hybrid
pv– thermal solar system for Cyprus. Renew Energy 2001;23:247 –60.
in Table 4. It can be seen from Table 4 that under this conserva-
[7] Tripanagnostopoulos Y, Nousia T, Souliotis M, et al. Hybrid photocoltaic/
tive assumption, the payback period is 14 years. Considering
thermal solar systems. Sol Energy 2002;72:217– 34.
that the cost (including water tanks, gas chambers and other [8] Krauter S. Increased electrical yield via water flow over the front of photo-
equipment) could be reduced with mass production and the voltaic panels. Sol Energy Mat Sol Cells 2004;82:131– 7.
additional rainwater collection can be recycled for domestic use [9] Abdolzadeh M, Ameri M. Improving the effectiveness of a photocoltaic
in non-operating period, the economic analysis results make this water pumping system by spraying water over the front of photovoltaic cells.
cooling approach quite attractive. Renew Energy 2009;34:91– 6.

International Journal of Low-Carbon Technologies 2014, 9, 118– 126 125

S. Wu and C. Xiong

[10] Kordzadeh A. The effects of nominal power of array and system head on the [16] Gang P, Jie J, Wei H, et al. Performance of photovoltaic solar assisted heat
operation of photovoltaic water pimping set with array surface covered by a pump system in typical climate zone. J Energy Environ 2007;6:1– 9.
film of water. Renew Energy 2010;35:1098 – 102. [17] Xu GY, Deng SM, Zhang XS, et al. Simulation of a photovoltaic/thermal
[11] Furushima K, Nawata Y. Performance evaluation of photovoltaic power gen- heat pump system having a modified collector/evaporator. Sol Energy
eration system equipped with a cooling device utilizing siphonage. Sol 2009;83:1967– 76.
Energy Eng 2006;128:146 –51. [18] Song B, Inaba H, Horibe A, et al. Heat, mass and momentum transfer of a
[12] Wilson E. Theoretical and operational thermal performance of a ‘wet’ crys- water film flowing down a tilted plate exposed to solar irradiation. Int J
talline silicon PV module under Jamaican condition. Renew Energy Therm Sci 1998;38:384– 97.
2009;34:1655 –60. [19] Smith CC, Lof G, Jones R. Measurement and analysis of evaporation from
[13] Akbarzadeh A, Wadowski T. Heat pipe-based cooling systems for photo- an inactive outdoor swimming pool. Sol Energy 1994;53 – 1:3– 7.
voltaic cells under concentrated solar radiation. Appl Therm Eng 1996;16: [20] Sutton Bonington. (2013) Weather data. http://www.metoffice.gov.uk/pub/
81 – 7. data/weather/uk/climate/stationdata/suttonboningtondata.txt. (Retrieved
[14] Huang MJ, Eames PC, Norton B. Thermal regulation of building-integrated on 3 April 2013).
photovoltaic using phase change materials. Int J Heat Mass Transf [21] Gould J. Rainwater catchment systems for domestic supply: design, construc-

Downloaded from https://academic.oup.com/ijlct/article-abstract/9/2/118/633328 by guest on 20 July 2020

2004;47:2715 –33. tion and implementation. Guildford: Biddles LtdGuildford, 1999, 45 – 68.
[15] Biwole P, Eclache P, Kuznik D. Improving the performance of solar panels [22] FITs. Feed-in tariff scheme. http://www.energysavingtrust.org.uk/
by the use of phase change materials, World Renewable Energy Congress Generating-energy/Getting-money-back/Feed-In-Tariffs-scheme-FITs (2013)
2011-Sweden. Photovoltaic Technol 2011;11:2953 – 60. (Retrieved on 25 April 2013).

126 International Journal of Low-Carbon Technologies 2014, 9, 118– 126