Applied Energy 88 (2011) 242–251

Contents lists available at ScienceDirect

Applied Energy
journal homepage: www.elsevier.com/locate/apenergy

Optimization and design of energy transport system for solar cooking application
U.R. Prasanna, L. Umanand *
Centre for Electronics Design and Technology, Indian Institute of Science, Bangalore 560 012, India

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

Article history: This paper proposes a hybrid solar cooking system where the solar energy is transported to the kitchen.
Received 16 April 2010 The thermal energy source is used to supplement the Liquefied Petroleum Gas (LPG) that is in common
Received in revised form 8 June 2010 use in kitchens. Solar energy is transferred to the kitchen by means of a circulating fluid. Energy collected
Accepted 21 July 2010
from sun is maximized by changing the flow rate dynamically. This paper proposes a concept of maxi-
Available online 24 August 2010
mum power point tracking (MPPT) for the solar thermal collector. The diameter of the pipe is selected
to optimize the overall energy transfer. Design and sizing of different components of the system are
Keywords:
explained. Concept of MPPT is validated with simulation and experimental results.
Hybrid solar cooking
Energy optimization
Ó 2010 Elsevier Ltd. All rights reserved.
Maximum power point tracking
Flow optimization
Energy transport
Liquefied Petroleum Gas (LPG)

1. Introduction Solar cooking with energy storage using pressurized water ves-
sel [10], phase change material [11], pebble bed thermal energy
Wood cut for cooking purpose contributes to the 16 million storage (TES) [12] and box-type solar cooker with auxiliary heating
hectares of forest destroyed annually, which is causing indoor air [13] have been proposed, which require the cook to work outdoors
pollution. The energy for cooking accounts for 36% of the total pri- in rural areas and on roof tops in urban areas. A split-system solar
mary energy consumption. The cooking energy demand in rural cooker exists, which has its flat-plate collector outdoors and the
areas of developing countries is largely met with bio-fuels such cooking chamber inside the kitchen, with heat pipes transferring
as fuel wood, charcoal, agricultural residues and dung cakes, the energy between the two [7,14]. Multipurpose solar cooker
whereas Liquefied Petroleum Gas (LPG) or electricity is predomi- cum water heater [15] and single basin solar still with PCM storage
nantly used in urban areas. Different energy sources for cooking [16] have been proposed. Heat exchanger for solar hot water stor-
have been evaluated in [1–4] and LPG stove is found to be the most age system [17] has been proposed. For a solar cooking system to
preferred cooking device in India. Solar cookers are expected to be accepted and adopted in most of the households, the following
contribute considerably towards meeting domestic cooking energy objectives have to be satisfied.
requirement in a country blessed with abundant sunshine [5].
Solar cooker is an environmental friendly and cost effective de- 1. The cooking should be done without moving out of the
vice for harnessing solar energy. The conventional box type cooker kitchens.
design has been studied and modified since 1980s and various de- 2. A reduction in the use of conventional energy.
signs and their characteristics have been extensively investigated 3. Cooking can be carried out at any time of day.
in [6]. Box type cooker [7] with multiple reflectors are easy to build 4. Time taken for cooking must be comparable with conventional
and use, but cooking has to be done outdoor and it is slow. Hot box cooking.
ovens [8] and concentrating solar cookers are cheap and effective;
however they are limited to cooking during clear sky periods. In order to satisfy the above mentioned objectives, a hybrid so-
Though parabolic cookers [9] are used for fast cooking, cooking rate lar cooking technique is proposed wherein the solar energy is
cannot be controlled and it is potentially hazardous due to focusing transferred to the kitchen and supplements the conventional LPG
of sun beam. source.
Consumption of conventional cooking fuel is minimized by
maximizing the energy obtained from sun. This maximization is
* Corresponding author. Tel.: +91 80 2360 0810; fax: +91 80 2293 2290. done by varying the flow rate of the fluid that is being circu-
E-mail address: [email protected] (L. Umanand). lated through the solar thermal collector. At lower flow rates,

0306-2619/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.apenergy.2010.07.020
 U.R. Prasanna, L. Umanand / Applied Energy 88 (2011) 242–251 243

Nomenclature

A surface area of the heat exchanger (m2) Powerpump power required by pump (W)
Aa aperture area (m2) Pr power received at the receiver (W)
Acoll area of collector (m2) Pu useful power (W)
Ao outer surface area of pipe (m2) P0u useful power per unit length (W/m)
APV area of PV panel (m2) Q maximum flow rate (m3/s)
Ar receiver area (m2) q_ flow rate (m3/s)
Cbattery capacity of battery (Ah) q_ 1 flow rate on collector side (m3/s)
Cp specific heat of the fluid (J/kg K) q_ 2 flow rate on load side (m3/s)
Cpf calorific value of the fuel Rb tilt factor for beam radiation
CR concentration ratio Re Reynold number
D diameter of pipe (m) Rd tilt factor for diffused radiation
DOD depth of discharge t thickness of insulation (m)
Do outer diameter (m) Tamb ambient temperature (K)
Di inner diameter (m) Tf temperature of fluid (K)
Dno-sun number of no-sunny days (day) Tfi temperature of the inlet fluid (K)
Ereq1 amount of energy required for cooking (kWh/day) Tfo temperature of the outlet fluid (K)
Estorage amount of energy to be stored (kWh) Thin temperature of the fluid inlet to heat exchanger (K)
F0 collector efficiency factor Thout temperature of the fluid outlet to heat exchanger (K)
FR heat removal factor Tmax maximum temperature of fluid in tank (K)
g earth’s gravitational constant (m/s2) Tmin minimum temperature of fluid in tank (K)
Gb beam or direct irradiation (W/m2) Tpipe temperature of pipe (K)
H pressure head (m) Tr temperature of the receiver (K)
Hd delivery head (m) Tsin temperature of the fluid inlet to storage tank (K)
Hf friction head (m) Tsout temperature of the fluid outlet to storage tank (K)
hfi heat transfer coefficient of absorber tube (W/m2 °C) u velocity of fluid (m/s)
Ho monthly averages of daily extra terrestrial global solar UL overall heat loss coefficient (W/m2 °C)
radiation (Wh/m2 day) Um overall thermal heat conductivity of heat exchanger (W/
Htc solar insolation at earth’s surface (Wh/m2 day) m2 °C)
Ib beam radiation (W/m2) V battery voltage (V)
Id diffused radiation (W/m2) Vf average fuel consumption per day
Io extra-terrestrial beam irradiance (W/m2) Vpump voltage applied to pump (V)
Ipump electric current drawn by pump (A) Vstorage volume of fluid in storage tank (m3)
Isc solar constant (W/m2) W width of the collector (m)
k thermal conductivity (W/m2 °C) Wh1 energy required from PV during daytime (Wh)
K1 fraction of daily load required for night cooking Wh2 energy required from PV during night (Wh)
K2 fraction of daily load required during no-sunny days Whbattery total energy required to store in battery (Wh)
k1 thermal conductivity of pipe (W/m2 °C) Whtotal total energy required from PV (Wh)
k2 thermal conductivity of thermal insulation (W/m2 °C) d declination angle (°)
kp absolute roughness of pipe material (m) DTm mean temperature difference between load and fluid (K)
kpump pump constant (m2/kg2) o emissivity of pipe surface
KT clearness index / latitude angle (°)
l length of pipe (m) g1 collector efficiency
L length of the collector (m) gB overall battery efficiency
m _ mass flow rate (kg/s) gbattery battery efficiency
mstorage mass of fluid in storage tank (kg) gc collector efficiency
N day of the year go optical efficiency
Pcoll collected power (W) gpump efficiency of pump
Pfl power flow from fluid to load (W) gPV efficiency of PV panel
Pin power available at input of the collector (W) gstorage efficiency of storage system
PL power loss to ambient from receiver (W) k Darcy friction factor
Ploss power loss to ambient from pipe (W) m kinematic viscosity (m2/s)
Ppipe hydraulic pressure drop across pipe (N/m2) q density of fluid (kg/m3)
Ppump power required by pump (W) r Stefan–Boltzmann constant (W/m2 K4)
Po power required to circulate fluid (W) xs hour angle at sunrise/sunset (°)

temperature of the collector and outlet fluid are higher resulting in This has lead to concept of maximum power point tracking
higher heat loss to ambient. Increasing the flow rate increases (MPPT).
the energy required for circulating the fluid, even though the Section 2 describes the proposed cooking system and its opera-
collected heat energy improves. For the given solar insolation tion. Optimization of energy by MPPT and selection of pipe diam-
and other external factors, there exists an optimal flow rate. eter are described in Section 3. Design of such a cooking system
Hence there is a need for variation in the flow rate dynamically is explained in Section 4. Results are presented in Section 5 and fi-
in order to optimize the energy absorbed from the collector. nally concluded in Section 6.
244 U.R. Prasanna, L. Umanand / Applied Energy 88 (2011) 242–251

2. Hybrid solar cooking of a receiver, where the sunlight is absorbed and converted to
required form of energy and a concentrator, which is an optical
2.1. System description system to concentrate direct radiation onto the receiver. Concen-
tration ratio CR is defined as the ratio of the area of aperture Aa
The block diagram of the proposed cooking system is as shown to the area of the receiver Ar.
in Fig. 1. The solar thermal collector is in general placed at a high Aa
location preferably on the roof top. A cylindrical (linear) parabolic CR ¼ ð1Þ
Ar
collector, a paraboloid or a concentrating collector is used to collect
solar energy and increase the temperature of the fluid. The heat ex- Temperature ranging from 100 °C up to 2000 °C can be obtained
changer is placed in the kitchen where the cooking is done. It by increasing the concentration ratio. Method to concentrate sun
transfers heat from the circulating fluid to the cooking load. All beam can be either by refraction as in case of fresnel lens or by
other components are placed at intermediate levels according to reflection. Reflection type of concentrating collectors are flat plate
the building requirements. Pump-I is used to vary the flow rate collector with plane mirror reflectors, linear (cylindrical) parabolic
of the fluid through the solar thermal collector. The energy ex- collector, compound parabolic collector, paraboloid collector etc.
tracted from the sun is stored in the buffer tank. The size of this Total power available at a concentrating collector is given by Eq.
tank is decided by the amount of energy that needs to be stored (2), where Gb is the beam or direct irradiance in W/m2.
for late night or early morning cooking and amount of energy that
Pin ¼ Gb Aa ð2Þ
needs to be saved from the other energy sources of the hybrid sys-
tem. Whenever food has to be cooked, the stored energy is trans- Only part of the power which is falling on the reflector, is con-
ferred to the load through the heat exchanger using pump-II, centrated onto the receiver depending on the optical efficiency go
which varies the flow rate of the fluid through the heat exchanger. given by (3). This efficiency depends on many factors like accuracy
The auxiliary source of energy like LPG or electrical energy is used of tracking mechanism, material and shape of the reflector [18]. Pr
for supplementing the stored solar energy and it as well reduces represents the amount of power available at the receiver.
the time required for cooking as compared to previously proposed
cooking systems like box-type cooker. Energy required from the Pr ¼ Gb go Aa ¼ go Pin ð3Þ
auxiliary source is to be optimized for the given system, availabil-
ity of solar insolation at the location and the load profile. Under steady state conditions, useful power Pu, obtained at the
receiver can be expressed as in Eq. (4). PL is the amount of power
2.2. System operation lost from the receiver to atmosphere through convection and radi-
ation, where UL is solar collector heat loss coefficient in W/m2 °C
The major goal of the proposed system is to transfer heat from and Tr is the temperature of the receiver in K.
the solar collector to the load. There are two levels of heat transfer
Pu ¼ Pr  PL ¼ Gb go Aa  U L ðT r  T amb ÞAr ð4Þ
with intermediate energy storage in a buffer tank. The heat is first
transferred from the solar collector to the storage tank. The pump-I The useful energy gain per unit length of the receiver can be ex-
controls the fluid flow rate q_ 1 to control the heat transfer from the pressed as,
collector to tank. Power from the auxiliary source, like LPG or elec-
trical heater is controlled according to the cooking load require- Pu Gb go Aa U L ðT r  T amb ÞAr
P0u ¼ ¼  ð5Þ
ment and availability of the stored energy and solar energy. L L L
Energy taken from this source has to be minimized so as to opti-
mize the savings of LPG. This useful power per unit length can be obtained in terms of
fluid temperature as given in Eq. (6).
 
3. Energy optimization F 0 Aa U L ðT r  T amb Þ
P0u ¼ Gb g o  ð6Þ
L CR
3.1. Solar heating
where F0 is called as collector efficiency factor. For collector having
0
For applications which require temperature above 100 °C, con- circular cross section with outer and inner diameter as Do and Di, F
centrating collectors are used. A concentrating collector comprises is given by (7) [19].

Fig. 1. Block diagram of the hybrid solar cooking system.

 U.R. Prasanna, L. Umanand / Applied Energy 88 (2011) 242–251 245

1 0.45
F0 ¼ h  i ð7Þ
50C
UL 1
þ hDDo i þ D2ko ln Do

Collector efficiency per unit

UL fi Di

where k is absorber thermal conductivity in W/m °C and hfi is heat 0.4

transfer coefficient inside absorbed tube in W/m2 °C. Useful power 100C

gained at the collector can be expressed in terms of inlet fluid tem- 0.35
perature Tfi and ambient temperature as in Eq. (8), where FR is the
heat removal factor. FR is a function of collector efficiency, specific
150C
heat of fluid, flow rate, dimension of collector, heat loss coefficient 0.3
UL etc. Eq. (8) is the equivalent of the Hottel–Whiller–Bliss equation
for liquid flat-plate collector [19,20].
0.25
Pu ¼ F R ½Gb go Aa  U L ðT fi  T amb ÞAr  ð8Þ
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2
Dividing the above equation by Pin, steady state collector efficiency
can be obtained as,
Mass flow rate in kg/s
  Fig. 2. Effect of mass flow rate on collector efficiency.
Pu Pr  PL U L ðT fi  T amb Þ
gc ¼ ¼ ¼ F R go  ð9Þ
Pin Pin Gb  CR
Applying first principle of thermodynamics to the solar collector
for open systems, useful power gain can also be written as rate of 0.46
specific heat gained by the fluid.
0.44
_ p DT f
Pu ¼ mC ð10Þ

Collector efficiency per unit

0.42
From Eq. (9), it is observed that the collector efficiency gc de-
pends mainly on two factors Pr and PL. Power reflected from the 0.4
concentrator can be optimized by improving the optical efficiency
go. This can be achieved by using reflector having better reflectivity 0.38
and shape. Once the reflector is fabricated, efficiency is fixed and
can not be improved further. 0.36
The other part of the collector efficiency is PL, which accounts
for radiative and convective heat loss from the receiver to atmo- 0.34
sphere. This is a function of the receiver temperature. Higher the
temperature, more is the heat loss. In order to increase the collec- 0.32
tor efficiency, the temperature of the receiver has to be kept as
close as possible to the ambient temperature in order to reduce 0.3
heat loss to ambient according to Eq. (5). 300 320 340 360 380 400 420
For a given constant inlet temperature of fluid and constant so- Collector Inlet Temperature in K
lar irradiance, as flow rate increases, the collector efficiency in-
Fig. 3. Effect of inlet temperature on collector efficiency.
creases as the collector outlet temperature comes down. Since
heat loss from collector to atmosphere is proportional to receiver Collector efficiency is calculated for different inlet temperature by
temperature, efficiency of collector is lesser at smaller fluid flow keeping mass flow rate at a fixed value. It can be observed from
rates. At very large flow rates, outlet temperature tends to inlet Fig. 3 that the collector efficiency increases as inlet temperature is
temperature according to Eq. (10), as input power is considered decreased toward ambient.
to remain same.
Performance of linear parabolic collector is found according to 3.2. Pumping power
the above equations. A MATLAB program is written which calcu-
lates steady state useful collected power and collector efficiency Power necessary for driving pump depends on type of the pump
analytically. Instantaneous collector efficiency is calculated for an that is being used for circulating fluid. Useful power Po that is being
inlet fluid temperature of 150 °C as in Eq. (11) where Ib and Id added to the fluid flow is given by,
are beam and diffused radiation, W is the width of the collector,
_ is the mass flow rate, Cp is the spe-
L is the length of the collector, m Po ¼ qgHq_ ð12Þ
cific heat of the fluid, Tfi and Tfo are the inlet and outlet temperature where q is the fluid density in kg/m , g is the acceleration due to 3

of the fluid. Tilt factor for beam and diffused radiation are repre- earth’s gravity, H is the head added to the flow in m and q_ is the flow
sented by Rb and Rd respectively. rate in m3/s. Input power necessary to drive the pump to deliver
_ p ðT fo  T fi Þ
mC pressure of DP is much more than the output power given by Eq.
gc ¼ ð11Þ (13), where the efficiency of the pump gpump is being considered.
ðIb Rb þ Id Rd ÞWL
DP  q_
Inlet temperature to the collector Tfi is varied by keeping the mass Ppump ¼ ð13Þ
flow rate constant. For lower inlet temperature, receiver tempera-
gpump
ture is less. Since heat loss PL is minimum, collector efficiency is For a centrifugal pump, input power is proportional to flow rate gi-
maximum. On the other hand, as inlet temperature is increased, re- ven by Eq. (14). kpump is a pump constant depending on the type and
ceiver temperature is more resulting in lower collector efficiency. design of the pump. As the mass flow rate increases, power required
For linear parabolic collector, inlet temperature is set at three differ- to circulate fluid through the collector increases drastically.
ent values (50 °C, 100 °C and 150 °C) and collector efficiency is cal-
culated for varied range of mass flow rate as shown in Fig. 2. _3
Ppump ¼ kpump m ð14Þ
246 U.R. Prasanna, L. Umanand / Applied Energy 88 (2011) 242–251

3.3. Effective collector efficiency 45

Effective Collector Efficiency in %

In an active mass transport solar system, additional energy is
spent on the supply-side pump to drive fluid through the collector 40
in order to extract energy from the sun. Hence efficiency of the sys-
tem containing solar collector and pump can be expressed as given
in Eq. (15). Ppump can be expressed in terms of Pu and equation can 35
be written, where Hf represents the friction head in the collector
side loop.

Effectiv e Collected Power 30

Effectiv e Collector Efficiency ¼
Input Solar Power
Pu  P pump 100%
¼ 25
Pin 80%
" # 40% 60%
Pu gHf
¼ 1 ð15Þ
Pin C p DT f gpump 20
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18
Initially, as the flow rate increases, power extracted from the Mass flow rate in kg/s
collector increases. After reaching some flow rate, power required
to pump the fluid increases more drastically as compared to in- Fig. 5. Effective collector efficiency for different insolations.

crease in collector power. Effective collector efficiency is calculated

from this collector power as given in Eq. (15). For linear parabolic
collector, the effective collector efficiency is analyzed for a con- MPPT control block diagram is shown in Fig. 6. Inlet tempera-
stant inlet temperature. Efficiency is plotted against mass flow rate ture Tsin and outlet temperature Tsout from the storage tank are
for different inlet temperatures as shown in Fig. 4. sensed using temperature sensors. Power collected from the solar
At lower flow rates, increase in collector efficiency is more com- collector is calculated according to the equation given by (16),
pared to increase in power required to pump the fluid. Effective where the mass flow rate m _ is sensed using a flow meter. Effective
collector efficiency increases at lower flow rates. Above some par- collector power which is calculated from Pcoll and Ppump, is fed to
ticular flow rate, pumping power increases and there is a flow rate the MPPT optimizing algorithm. This gives duty ratio as output,
at which power extracted from solar collector is maximum. In or- which in turn controls the voltage level of the supply given to
der to maximize the collector efficiency, one has to operate at this the pump through the power converter as shown in Fig. 6.
peak point.
_ p ðT sin  T sout Þ
Pcoll ¼ mC ð16Þ
3.4. Maximum power point tracking (MPPT)
MPPT algorithm can be grouped into two major types, namely
Performance analysis is carried out for different solar insolation direct and indirect methods [21]. Indirect methods track the peak
level at constant ambient air temperature and inlet thermo-con- power from the measures of the system like temperature, irradi-
vector fluid temperature. From Fig. 5, it can be observed that for ance or using empirical data or lookup table, by mathematical
100% solar insolation, optimal mass flow rate is around 0.04– expressions through numerical approximations. For this, charac-
0.05 kg/s. As solar insolation is reduced, the optimal flow rate shifts teristic of the system has to be known before hand and it is specific
towards lower value. This shows that the optimal flow rate is not only to that system. Hence, they do not track maximum power for
fixed for a given system. It is a function of location and solar inso- any given external conditions exactly.
lation at that particular instant. Hence one has to vary the flow rate On the other hand, direct method calculates the actual power
accordingly in order to extract optimal energy from sun. This leads absorbed from the sun and tries to maximize it. These type of algo-
to the concept of maximum power point tracking also known as rithms are suitable for any irradiance and temperature. Most com-
MPPT. monly used MPPT algorithm is perturbation and observation
method [22]. This is an iterative method of obtaining MPP. This
measures the solar collector characteristics and then perturbs the
operative point to know the direction of change in power. Maxi-
50 mum point is reached when change in power collected is zero
Effective Collector Efficiency in %

for a small change in the flow rate. Amount of the perturbation gi-
40 ven is calculated according to the slope of the present operating
point. This method has advantage over other methods as it can
30
be applied to any system without prior knowledge about its actual
characteristic.
20
Flow chart for implementing MPPT is as shown in Fig. 7. Power
10 obtained from the collector and power supplied to the pump are
50C
calculated. Rate of change in effective collector power is estimated.
0 Accordingly the value of duty ratio is calculated. Duty ratio is
150C 100C either increased or decreased depending on the present operating
−10 point. If the operating point is on the left side of the inverted ‘U’
curve (Fig. 5), then change in duty ratio and change in power are
−20 of same sign. Hence duty ratio is increased such that operating
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2
Mass flow rate in kg/s point moves toward optimal value. On the other hand, change in
duty ratio and power is in opposite direction when operating right
Fig. 4. Effective collector efficiency for different inlet temperatures. side of optimal flow rate. The gain value K is decided according to
 U.R. Prasanna, L. Umanand / Applied Energy 88 (2011) 242–251 247

Fig. 6. Control block diagram for MPPT.

The characteristic equation for a heat exchanger is given by Eq.

(17). UmA gives the mean overall thermal conductivity between the
heat exchanger and the load. This is decided by the dimension,
shape of the heat exchanger and conductivity of the material used.

Pfl ¼ U m ADT m ð17Þ

where Pfl is the heat flow from fluid to load, Um is the mean overall
heat transfer coefficient, A is the total surface area of the load heat
exchanger and Tm is the mean temperature difference between fluid
and load. It is observed from Eq. (17) that DTm should be maximized
to increase the heat transfer rate. For a given temperature difference
DTm, energy flow can be increased by maximizing the conductive
barrier between the hot fluid and the cooking load. Design should
be done such that the energy transferred to the load should be max-
imum with minimal pressure drop across it. DTm can be maximized
by taking the fluid from top of the storage tank which has the high-
est temperature due to density stratification. Once the heat energy
is transferred to the load, the lower temperature fluid is fed back to
the buffer tank.
The above equation can also be written as,

_ p ðT hin  T hout Þ
Pfl ¼ mC ð18Þ

where m _ is the mass flow rate through the heat exchanger, Cp is the
specific heat of the fluid, Thin and Thout are the temperatures of fluid
entering and leaving the heat exchanger respectively. Power trans-
ferred to the load can be controlled by varying the mass flow rate m. _
Pump-II is supplied by a variable power supply unit by which the
flow rate can be controlled. When power supplied from the storage
tank is not sufficient, the auxiliary source of energy like LPG or elec-
trical energy is used.
LPG and solar energy are integrated in such a way that, both of
them heat up only the load and are mutually exclusive. During the
condition when both energy sources are supplying heat, the tem-
Fig. 7. Flowchart for MPPT algorithm. perature of the heat-transfer fluid is higher than load. In such a
case the entropy flow is to the load (sink). If the temperature of
heat-transfer fluid is lower (during night or no-sun conditions),
the piping structure is such that natural thermosyphon of the ther-
the system dynamics like time constant of the system and varia- mic fluid from the load side to buffer tank does not exist. Further
tion in solar insolation. isolation is provided by two insulating isolation valves that isolate
the thermic fluid at the load exchanger and the buffer tank.
3.5. Load-side optimization
3.6. Selection of pipe diameter
The other part of the system transfers heat energy from the buf-
fer tank to the load which is placed in the kitchen. The buffer tank Performance of the solar system is dependent on the pipe diam-
is used to store the heat energy in the form of sensible heat, which eter used for circulation of the fluid. There are two contradictory
is used for cooking. With this storage unit, cooking can be carried effects, which are dependent on diameter of the pipe. One of them
out even when energy from sun is not available. The heat energy is the heat loss to ambient, which varies directly with variation in
is transferred to the load by circulating the hot fluid stored in the diameter. Another effect is the pumping power required to circu-
buffer tank through a heat exchanger. The pump controls the flow late the fluid, which increases as diameter is decreased. These
rate in this loop thereby controlling the rate of heat transferred to two contradicting effects lead to the existence of an optimal pipe
the load. diameter for a given system.
248 U.R. Prasanna, L. Umanand / Applied Energy 88 (2011) 242–251

3.6.1. Pressure drop 450

Consider a pipe of length L having a constant diameter D. Pres-
sure drop across this pipe is given by the Darcy–Weisbach formula 400
[23] as in equation.
350
l qu2
Ppipe ¼k ð19Þ 300
D 2

Power in W
Ploss+Ppump
where u is the velocity of the fluid in m/s and k is a dimensionless 250
coefficient called the Darcy friction factor or coefficient of line
hydraulic resistance, which can be found from a Moody diagram 200
[23]. In case of laminar flow of fluids, the Darcy factor is calculated 150 Ploss
using Stokes formula [23] given by,
100
64
k¼ ð20Þ
Re 50 Ppump

where Re is the Reynold Number, given by

0
Du 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018
Re ¼
m
ð21Þ Diameter in m

where m is the kinematic viscosity of the fluid. By substituting Eqs. Fig. 8. Effect of diameter on the performance.
(20) and (21) in Eq. (19), the pressure drop along a pipe is obtained
as,
l 64mlm
_ becomes equal to heat transferred from pipe to ambient. For a
Ppipe ¼ 32mq 2
u¼ ð22Þ given value of fluid and ambient temperatures and pipe tempera-
D 2 p D4
4 ture Tpipe, the power loss Ploss can be calculated by solving the
In most of the solar applications, flow rate is very small and non-linear Eqs. (24) and (25). It is observed from the above equa-
Reynolds number Re is less than 2000. Hence the flow is considered tions that the heat loss to ambient is directly proportional to the
as laminar. If flow is turbulent, the friction factor k is calculated surface area of the pipe. Hence heat energy lost is higher if larger
according to a fit given by Colebrook [24], diameter pipes are used.
  Discussions in Sections 3.6.1 and 3.6.2 show that Ppump and Ploss
1 2:51 0:269kp change inversely as the pipe diameter is varied. Hence there exists
pffiffiffi ¼ 2log pffiffiffi þ ð23Þ
k Re k D an optimal pipe diameter for which Ppump + Ploss is minimal. For a
where kp is the absolute roughness of the pipe material used. For given solar system, diameter of the pipe is selected in order to min-
both laminar and turbulent flow, as the diameter of the pipe de- imize the sum of these two power losses. Optimal diameter of the
creases, the pressure drop increases. This increases the burden on pipe is obtained by solving the Eq. (26).
the pump which is circulating the fluid. For a given flow rate, higher @Ppump @Ploss
power is necessary to pump the fluid through a smaller diameter ¼ ð26Þ
@D @D
pipe.
Ppump and Ploss are calculated for a typical set of specifications. Var-
iation in these powers are plotted with change in diameter of the
3.6.2. Heat loss
pipe. Fig. 8 shows that as the diameter increases, Ploss increases
Heat energy is transferred from collector to storage tank and
whereas Ppump decreases. From the plot of Ploss + Ppump, it is observed
from storage tank to heat exchanger by circulating heat transfer
that there exists an optimal diameter at 8.26 mm for which the total
fluid through the pipe. Since temperature of the fluid is much high-
power loss is minimum.
er than ambient temperature, the fluid looses heat energy to ambi-
ent. Consider a pipe of length l carrying fluid having temperature Tf.
Heat is lost from fluid by conduction through the pipe wall given 4. Design of solar cooking system
by (24), where inner and outer diameter of the pipe are repre-
sented as Di and Do and k1 is the conductivity of the pipe material In a practical cooking system, there are different components
[25]. whose design issues are to be addressed considering the worst case
conditions for a given location and cooking load.
2pk1 l
Ploss ¼   ðT f  T pipe Þ ð24Þ
ln Do 4.1. Estimation of solar collector size
Di

This heat energy raises the temperature of the pipe. Heat from This mainly depends on the availability of solar energy at an
pipe is lost to ambient through the insulation by conduction and user defined location and the amount of energy required for cook-
radiation given by Cengel [25], ing. The monthly averages of daily extra terrestrial global solar
  radiation for a horizontal surface Ho at any location can be calcu-
2p k l
Ploss ¼ h 2 i ðT pipe  T amb Þ þ Ao ro T 4pipe  T 4amb ð25Þ lated by Eq. (27).
ðDo þ2tÞ
ln Do 24
Ho ¼ Io ½sinð/ÞsinðdÞxs þ cosð/ÞcosðdÞsinðxs Þ Wh=m2 day ð27Þ
where Ao is the outer surface area of the pipe, k2 is the conductivity p
of the insulation material, t is the thickness of the insulation, r is where
the Stefan–Boltzmann constant and o is the emissivity of the outer
surface. Io = Extra-terrestrial beam normal irradiance on a day in W/m2
 

Under steady state, all the temperatures reach an equilibrium = Isc 1 þ 0:033cos 360N
365
value. Under this condition heat transferred from fluid to pipe Isc = Solar constant = 1367 W/m2
 U.R. Prasanna, L. Umanand / Applied Energy 88 (2011) 242–251 249

N = Day of the year (N = 1 on January 1st and N = 365 on Decem- sizing of tank during no-sunny days. From Eq. (33), the size of the
ber 31st) storage tank can be calculated as,
/ = Latitude of the location in degrees h i
d = Declination angle in degrees = 23:45sin 360ðN80Þ Estorage ¼ mstorage C p ðT max  T min Þgstorage
365
1
xs = Hour angle at sunrise/sunset = cos (tan(/)tan(d)) Estorage
mstorage ¼
C p ðT max  T min Þgstorage ð34Þ
A stochastic measure of atmospheric effects is called as clear- mstorage
V storage ¼
ness index KT, which is a periodic function of time of year. The glo- q
bal solar insolation with atmospheric effects is obtained as,
where mstorage is the mass of fluid in the storage tank, Vstorage is the
Htc ¼ K T Ho ð28Þ volume of the tank, gstorage is the efficiency of storing and retrieving
the energy from the tank. Tmax and Tmin are the maximum and min-
where Htc represents the solar insolation at the earth’s surface on a
imum temperature of the fluid in the tank.
horizontal plate at any location on any given day. The monthly aver-
age daily insolation is calculated for all the months over a year con-
sidering a typical day of each month. From these values, the lowest 4.3. Sizing of battery
value of the solar insolation is considered for the design of solar
thermal and photo-voltaic collectors. Battery is used to meet the energy requirement from the electri-
According to the type and the amount of cooking fuel used per cal components during the night time. Total energy that needs to
day, the average energy requirement can be estimated according to be stored in the battery Whbattery is calculated by adding individual
Eq. (29). The Average fuel consumption for cooking per day, Vf is Wh required during night time and no-sunny day.
multiplied by the calorific value Cpf of that fuel to obtain the X
amount of energy required in kWh/day. Whbattery ¼ Wh
X Whbattery ð35Þ
Ereq1 ¼ ðV fi C pfi Þ8 i ð29Þ C battery ¼
i V  DOD  gbattery

Considering g1 as the efficiency of the collector that is being where Cbattery is the capacity of the battery in Ah, DOD is depth of
used, area of the collector in m2 can be calculated as in Eq. (30). discharge, V is the battery voltage and gbattery denotes the efficiency
Ereq1 of the battery.
Acoll ¼ ð30Þ
g1 Htc
4.4. Sizing of pumps
where Htc is as given in Eq. (28).
In the proposed system, photo-voltaic (PV) panels are used to
Two pumps are used in the proposed system, one to circulate
supply electrical energy requirement by pumps, linear actuator,
the fluid through the solar thermal collector and another through
sensing instruments like temperature sensor and flow meter
the heat exchanger. During steady flow condition in a closed cir-
[26]. Let Wh1 be the total energy required during daytime per
cuit, pump has to supply only friction head. In order to start the
day represented in kWh. This includes energy required by linear
flow, pump has to circulate the fluid against corresponding
actuator, pump-I, pump-II and other monitoring and controlling
delivery head. Hence rating of the pump depends on the static
instruments. Let Wh2 be the total energy required during night-
head to which the fluid has to be pumped and the flow rate.
time. This involves supply of power to mainly load-side pump
Pressure head and power rating of the pumps can be calculated
(pump-II), monitoring and control on load-side. Total energy re-
by Eq. (36).
quired to be supplied from the PV panel is given by,
9
Wh2 ðWh1 þ Wh2 Þ  Dno-sun Ppumpi ¼ qgHdi inPa =
Whtotal ¼ Wh1 þ þ For i ¼ 1; 2: ð36Þ
gB gB Powerpumpi ¼
Ppumpi Q i
inW ;
  gpump
Dno-sun Wh2
¼ Wh1 1 þ þ ð1 þ Dno-sun Þ ð31Þ
gB gB where Ppumpi is the pressure head, Powerpumpi is the power rating of
the pump, Hdi is the delivery head in m, gpump is the efficiency of the
where gB is the overall battery efficiency and Dno-sun is the number
pump and Qi is the maximum volume flow rate in m3/s.
of no-sunny days. Knowing insolation Htc as calculated in Eq. (28),
the area of PV needed to fulfill the electrical loads is obtained as gi-
ven in Eq. (32), where gPV is the efficiency of the PV panel. 4.5. Heat exchanger
Whtotal
APV ¼ ð32Þ In the proposed system, the heat exchanger is used to transfer
Htc  gPV heat from the fluid at higher temperature to the food which is
being cooked. Efficiency of heat transfer has to be maximum and
4.2. Sizing of the storage tank pressure drop across the exchanger has to be minimized. Size of
the heat exchanger has to be flexible with the dimension of the
Heat storage tank stores solar energy in the form of heat by rais- cooking vessel that is being used.
ing the temperature of the fluid. The amount of energy to be stored
depends on the amount of night cooking load and the number of 4.6. Monitoring and control system
no-sunny days given by Eq. (33).

Estorage ¼ ½K 1 Ereq1 þ K 2 Ereq1  Dno-sun  ð33Þ Parameters like temperature and flow rate are monitored plac-
ing temperature sensors at different points and flow meter on both
where K1 is the fraction of daily load required for night cooking and the collector side and load side. Control circuits are designed to
K2 is the fraction of cooking load that needs to be considered for track the sun, maximum power point etc.
250 U.R. Prasanna, L. Umanand / Applied Energy 88 (2011) 242–251

5. Experimental results

Fig. 9 shows the block diagram of the experimental setup for so-
lar cooking system. Paraboloid dish concentrator is used to focus
sun rays onto the receiver. Aluminium sheets are used as reflecting
material. To improve optical efficiency, surface of the reflector is
anodized. A linear actuator is fixed to the paraboloid with a lever
system in such a way that when actuator moves to and fro, the
paraboloid is rotated in east–west direction. Using an accelerome-
ter sensor that is fixed on the paraboloid, the tilt angle is sensed.
Early morning, the concentrator is fixed toward sun manually.
The sensor considers this as the reference angle and tracks at a
constant rate of 15° angle per hour.
A coil made of copper tube is placed at the focus of the parab-
ola in order to receive the heat. Servo-therm oil is circulated
through the collector to absorb heat energy. Stainless steel pipes
with glass wool insulation over that is used for circulation of the
oil from the collector to the tank. Thermocouples are placed to
measure temperature of oil entering and leaving the receiver. A
rotary pump is used to circulate oil through the receiver and
put back into the heat storage tank. The pumps are driven by per-
manent magnet DC (PMDC) motors, which are controlled by var-
iable voltage power supplies. The heat storage tank is made of Fig. 10. Experimental result showing MPP.
stainless steel material with good thermal insulation around that
for better retention of heat. The flow rate of the fluid is measured
using a flow meter. shown in Fig. 10. This clearly shows the optimal flow rate at which
On the load-side, hot oil from top of the heat storage tank is ta- the collected power is maximum.
ken to the kitchen through thermally insulated stainless steel tube.
Heat is transferred from oil at higher temperature to cooking load
using a heat exchanger. Helical shaped coil of copper is wound 6. Conclusion
around the cooking vessel with thermal insulation to constrain
the heat within the heat exchanger. Oil leaving the heat exchanger This paper proposes the hybrid solar cooking system, wherein
is pumped back from the kitchen to the buffer tank using another the solar energy is transferred to the kitchen and supplements
similar pump-motor drive. The mass flow rate on the load-side is the conventional LPG source. Cooking can be carried out at any
measured using another flow meter. Thermocouples (TC) are time of the day with time taken being comparable to conventional
placed at different places as shown in figure to measure the tem- systems. Energy collected from the solar thermal collector is opti-
peratures at various points of the system. mized by dynamically varying the flow rate using maximum power
In order to show the maximum power point, an electrical heater point tracking technique. Effect of variation in diameter of the pipe
is used in place of solar thermal collector to achieve control over is analyzed and method to select optimal diameter is proposed in
input power. For the same external conditions like temperature order to maximize the efficiency of the system. Design and sizing
of storage tank and input power, experiment is repeated for differ- of different components of the system are described with equa-
ent flow rate. Collected power and power supplied to the pump are tions. Simulation and experimental results are presented showing
calculated. Effective collector power is plotted against flow rate as the MPP.

Fig. 9. Block diagram of the experimental setup.

 U.R. Prasanna, L. Umanand / Applied Energy 88 (2011) 242–251 251

References [14] Chow TT, Pei G, Fong KF, Lin Z, Chan ALS, Hea M. Modeling and application of
direct-expansion solar-assisted heat pump for water heating in subtropical
hong kong. Appl Energy 2010;87:643–9.
[1] Ramanathan R, Ganesh LS. A multi-objective analysis of cooking-energy
[15] Kumar Naveen, Chavda Tilak, Mistry HN. A truncated pyramid non-tracking
alternatives. Energy 1994;19(4):469–78.
type multipurpose domestic solar cooker/hot water system. Appl Energy
[2] Pohekar SD, Ramachandran M. Utility assessment of parabolic solar cooker as
2010;87:471–7.
domestic cooking device in India. Renew Energy 2006;31:1827–38.
[16] El-Sebaii AA, Al-Ghamdi AA, Al-Hazmi FS, Faidah Adel S. Thermal performance
[3] Pohekar SD, Kumar Dinesh, Ramachandran M. Dissemination of cooking
of a single basin solar still with pcm as a storage medium. Appl Energy
energy alternatives in India – a review. Renew Sust Energy Rev 2005;9:379–93.
2009;86:1187–95.
[4] Ghafghazi S, Sowlati T, Sokhansanj S, Melin S. A multicriteria approach to
[17] Mondol Jayanta Deb, Smyth Mervyn, Zacharopoulos Aggelos, Hyde Trevor.
evaluate district heating system options. Appl Energy 2010;87:1134–40.
Experimental performance evaluation of a novel heat exchanger for a solar hot
[5] Purohit P, Kumar A, Rana S, Kandpal TC. Using renewable energy technologies
water storage system. Appl Energy 2010;86:1492–505.
for domestic cooking in India: a methodology for potential estimation. Renew
[18] Odeh SD, Morrison GL, Behnia M. Modelling of parabolic trough direct steam
Energy 2002;26:235–46.
generation solar collectors. Solar Energy 1998;62(6):395–406.
[6] Schwarzer Klemens, Vieira da Silva Maria Eugenia. Characterisation and design
[19] Sukhatme SP. Solar energy: principles of thermal collection and storage. TMH
methods of solar cookers. Solar Energy 2007;82(4):157–63.
Publishing Company; 2003.
[7] Khalifa AMA, Taha MMA, Mannaa A, Akyrt M. A split-system solar cooker with
[20] John Twidell, Tony Weir. Renewable energy resources. Taylor-Francis; 2008.
heat pipes. Energy Convers Manage 1996;26(2):259–64.
[21] Salas V, Olias E, Barrado A, Lazaro A. Review of the maximum power point
[8] Garg HP, Bandyopadhyay B, Datta Gouri. Mathematical modelling of the
tracking algorithms for stand-alone photovoltaic systems. Solar Energy Mater
performance of a solar cooker. Appl Energy 1983;14:233–9.
Solar Cells 2006;90:1555–78.
[9] Al-Soud Mohammed S, Abdallah Essam, Akayleh Ali, Abdallah Salah, Hrayshat
[22] Al-Atrash H, Batarseh I, Rustom K. Statistical modeling of dsp-based hill-
Eyad S. A parabolic solar cooker with automatic two axes sun tracking system.
climbing mppt algorithms in noisy environments. Appl Power Electron Conf
Appl Energy 2010;87:463–70.
Expos 2005;3:1773–7.
[10] Morrison GL, Di J, Mills DR. Development of a solar thermal cooking system.
[23] Larock Bruce E, Jeppson Roland W, Watters Gary Z. Hydraulics of pipeline
FMT/1 Report; 1993.
systems. USA: CRC Press; 2000.
[11] Buddhi D, Sharma SD, Sharma Atul. Thermal performance evaluation of a
[24] Colebrook CF. Turbulent flow in pipes. J Inst Civil Eng 1938;11:133.
latent heat storage unit for late evening cooking in a solar cooker having three
[25] Cengel Yugnus A. Heat transfer – a practical approach. McGraw-Hill
reflector. Energy Convers Manage 2003;44:809–17.
Professional; 2003.
[12] Mawire A, McPherson M, van den Heetkamp RRJ, Mlatho SJP. Simulated
[26] Ramchandra Bhandari, Ingo Stadler. Electrification using solar photovoltaic
performance of storage materials for pebble bed thermal energy storage (tes)
systems in nepal. Appl Energy 2009. doi:10.1016/j.apenergy.2009.11.029.
systems. Appl Energy 2009;86:1246–52.
[13] Hussain M, Das KC, Huda A. The performance of a box-type solar cooker with
auxiliary heating. Renew Energy 1997;12(2):151–5.