Energy Conversion and Management 150 (2017) 451–462

Contents lists available at ScienceDirect

Energy Conversion and Management

journal homepage: www.elsevier.com/locate/enconman

Performance parameter evaluation, materials selection, solar radiation with MARK

energy losses, energy storage and turbine design procedure for a pilot scale
solar updraft tower
⁎
Ramakrishna Balijepalli, V.P. Chandramohan , K. Kirankumar
Department of Mechanical Engineering, National Institute of Technology Warangal, Warangal, Telangana State 506 004, India

A R T I C L E I N F O A B S T R A C T

Keywords: Performance investigation of a prototype of solar updraft tower (SUT) power plant is carried out in this study.
Solar updraft tower The diameters of the solar collector and chimney are of the 3.5 and 0.6 m respectively. The objective of this work
Chimney is to investigate the performance of SUT and to tabulate all the inputs and estimated parameters with the
Solar collector materials of a SUT power plant. All the three main components’ (turbine, solar collector and a chimney) process
Turbine design
parameters are estimated and discussed. Appropriate materials are discussed and selected for solar collector,
Beam and diffuse radiation
chimney, turbine and heat storage materials. Solar beam, diffuse and global radiation are estimated to analyze
Energy storage materials
Selection of materials the performance of collector cover. Energy losses in solar collector cover and transmissivity estimations are
performed to calculate theoretical energy collected in solar collector. Pressure drop inside the chimney is esti-
mated and from that the actual power output of the turbine is calculated. The quantity of heat storage materials
needed is evaluated in terms of both mass and volume. Theoretically the maximum velocity of air is achieved at
the chimney base and is 2 m/s. The maximum overall efficiency of the plant is estimated as 0.0028%. The
maximum theoretical power output of the plant is 0.633 W.

1. Introduction
Power generation using solar energy is an attractive alternate so-
lution nowadays. There are two ways the power can be generated by
solar energy: through photo-voltaic cells and by solar thermal gen-
erator. Solar updraft tower (SUT, also called solar chimney power
plant) is also one of the key solutions to the today’s predominant energy
challenges. Unlike conventional power generating stations (such as
thermal power plants), solar chimneys do not require cooling water. A
solar chimney consists of three main components, namely, solar col-
lector, a chimney and a power generating unit – turbine.
Solar chimney converts only a small percentage of the collected
solar energy into electricity. However it can be overcome this dis-
advantage by making a cheap, robust construction with low main-
tenance costs [1,2]. The concept of solar chimney technology was first
introduced in 1931 [2] and after consequent studies the first Spanish
prototype SUT having a chimney height (Hch) of 195 m and a maximum
electricity generation of 50 kW was built and commissioned in Man-
zanares, Spain in 1982. This work concluded that effective electricity
production is possible with the application of large scale (up to
400 MW) solar power plants [2]. Once this setup was made then the
electricity can be produced without any further cost as it has no running
cost. One of the other important fact is that it is eco-friendly com-
parative to other fossil fuel plants. These factors inspired to develop this
complete data of SUT.
Preliminary experiments were carried out by Haaf [3] in a solar
chimney power plant. This work discussed and analyzed about the
thermal energy balances, solar collector efficiency, various pressure
losses and turbine losses with reference to a twenty four hour period.
The current status of knowledge such as physical process, experimental
and theoretical study and cost evaluation for the solar chimney power
plant has been reviewed by Zhou et al. [4]. Descriptions of other types
of solar chimney power technology were also mentioned in this work.
The physical process of a large scale SUT was evaluated by Pretorius
and Kröger [5]. This work incorporated the new convective heat
transfer coefficient correlation and found that the annual power output
of the system reduced up to 11.7%. It was also found that the annual
power output increases by 3.4% by selecting better glass material for
collector. Simulations were conducted and it was concluded that the
results for energy storage materials such as lime stone and sand stone
gave similar results to that obtained for granite.
A 2-D numerical model was developed by Shrivan et al. [6] to

⁎
Corresponding author.
E-mail address: vpcm80@gmail.com (V.P. Chandramohan).

http://dx.doi.org/10.1016/j.enconman.2017.08.043
Received 18 May 2017; Received in revised form 24 July 2017; Accepted 15 August 2017
0196-8904/ © 2017 Elsevier Ltd. All rights reserved.
R. Balijepalli et al. Energy Conversion and Management 150 (2017) 451–462

Nomenclature Vr volume of the rocks (m3)

w relative wind speed (m/s)
A solar azimuth angle (°) Z solar zenith angle (°)
Ac collector surface area (m2)
b width of the blade section (m) Symbols
B number of blades
Copt chord length (m) α angle of attack (°)
Cp specific heat of the air (kJ/kg K) β turbine blade pitch angle (°)
CL coefficient of lift ξ collector slope angle (°)
CD coefficient of drag δ solar declination (°)
D diameter (m) εA absorber plate emittance
DY the day of the year εg collector cover emittance
FL lift force (N) λ turbine blade tip speed ratio.
FD drag force (N) υ kinematic viscosity of air (m2/s)
g acceleration due to gravity (m/s2) ρ reflectance
H distance between the absorber plate and the collector ρa density of the air (kg/m3)
cover (m) Ψ the azimuthal collector orientation (°)
hw wind loss coefficient (W/m2 K) σ Stephan-Boltzmann constant (W/m2 K4)
hL linearized radiation coefficient (W/m2 K) σs solidity ratio
hc convective heat transfer coefficient (W/m2 K) θ the obliquity angle (°)
H height (m) τ the clearness index
K thermal conductivity (W/m·K) ω rotational speed of the turbine (rpm)
L latitude (°) ωs the hour angle at sunset and sunrise (°)
L′ colatitude (°)
mr mass of the rocks (kg) Subscripts
n number of days after the vernal equinox
N number of rotations per minute (rpm) a ambient
r radius of blade segment from the hub (m) A absorber plate
R turbine blade tip radius (m) c cover
S the solar constant (W/m2) ch chimney
T temperature (°C) g glass transition
Ut overall heat loss coefficient (W/m2·K) h hot air
v velocity (m/s) m glass melting
v1 free stream velocity (m/s) max maximum

estimate the maximum power output from the SUT power plant. The
design data were used from the recently erupted plant from Iran. The
effect of entrance gap of the collector, chimney diameter, chimney
height and inclination of collector roof on maximum power output was
studied. The optimized parameters were identified through this nu-
merical work.
Tian and Zhao [7] discussed the most recent growths and ad-
vancements in solar thermal energy applications. A brief review has
been done over solar collectors and various thermal energy storage
systems. They concluded that photovoltaic thermal collectors gave the
best overall performance among the various non-concentrating type
collectors. This work also suggested that the molten salts are the best
choice for high temperature thermal energy storage applications be-
cause of their wide properties such as density, thermal conductivity and
specific heat.
A similar type of overview is given by Schlaich et al. [8]. This work
described the price evaluation for a large scale solar chimney power
plant. It explained the results of a newly designed power plant in Spain.
Technical and economic issues which were related to the upcoming
commercial solar power plants were also discussed. An experimental
investigation was carried to check the functioning of a small scale solar
tower turbine by von Backström and Gannon [9]. The inlet guide vanes
provided in their setup were meant to enhance the pre-whirl of air
which reduces the turbine exit kinetic energy at diffuser inlet. The
turbine developed in the setup produced 85–90% total-to-total effi-
ciency and 87–90% total-to-static efficiency.
The performance of different airfoil structures such as symmetric
and cambered of standard three blade vertical axis wind turbine was
examined by Durrani et al. [10] through CFD numerical simulation. The
impact of thickness of blade, camber, turbulent flow intensity and the
effect of various turbulence models were discussed for acquiring the
overall most beneficial turbine blade design configuration. Similarly a
mathematical model for SUT was developed by Gitan et al. [11] and
Guo et al. [12] and they found the optimized slope angle of a tilted
collector. The mathematical model was compared with an experimental
result and good agreement was noticed.
A divergent chimney was proposed for SUT power plant by Hu et al.
[13]. A numerical model was developed with different configuration of
divergent chimney with other variable parameters of area ratio and
divergent angle. They concluded that the performance of the divergent
chimney was much higher than the straight chimney. Similarly a
mathematical model was developed for the performance analysis of
SUT by Maghrebi et al. [14] for a sloped chimney.
An elaborated experimental study was carried out by Ozgen et.al
[15] in order to investigate the thermal performance of three categories
of double flow solar air heaters having aluminium cans under different
operating conditions. They concluded that the double flow air channels
are more efficient than the single flow channel over or under the ab-
sorber plate. Similar studies have been performed with artificial neural
networks (ANN), wavelet neural network (WNN) [16] and least squares
support vector machine (LS-SVM) [17]. Finally it was concluded that
WNN was an excellent method for prediction of thermal performance of
the system. Similar investigation of greenhouse heating using various
renewable energy sources under different climatic conditions was re-
ported by Esen and Yuksel [18].
From the literature survey it is noticed that quite a few prototypes of
SUT power plant have been developed [8,9]. Some studies explained
the SUT components such as turbine [15–22] and chimney [2,3]. In few

452
R. Balijepalli et al.
of the works CFD simulations were performed [6,10–12,20] and an
analytical solution [23] was used to identify the optimum parameters of
turbine. Few studies reviewed [4,24] the technical and economic issues
of SUT. Overall, this setup has been enormously reviewed, though there
is a dearth of design parameters, as this setup involves a number of
process and design parameters, such as turbine blade parameters, solar
radiation parameters, quantity and selection of storage materials, ma-
terials for solar collector (cover and absorber plate). No performance
and design details of SUT have been found for the main components
such as solar collector, chimney and turbine. The solar input parameters
have not been estimated for this specific application and subsequent
energy losses also not found. Similarly, the materials needed for the
above components have not been discussed. There is no solar energy
storage facility found as the system has a huge area of solar collector.
Therefore, a step is taken in this work to address the above issues in a
very clear manner. The present setup is considered with an energy
storage option with all estimation procedures and values tabulated.
Therefore, the main objectives of this work are (i) to do a perfor-
mance investigation of a solar chimney power plant components such
as chimney, turbine, solar collector and to perform design calculation
for turbine, (ii) to estimate the pressure loss calculations, thermal ra-
diation and heat loss calculations for finalizing the optimized design
and process parameters such as required collector surface area, power
output of total system and efficiency. (iii) To select appropriate mate-
rials for collector, absorber plate, chimney, turbine, supports and
thermal energy storage system, (iv) to determine the amount of storage
material required to store the energy for night use as well as energy
storage for overcast days and (v) to tabulate all design and performance
parameters of SUT, so a complete data, performance and process
parameters of SUT can be given to future researchers.
2. Methodology: description and design
The setup consists of solar collectors, chimney and an electricity
generating turbine as shown in Fig. 1. In day time, radiation from the
sun reaches the cover of solar collector where a specific amount of
energy is reflected, some amount of energy is transmitted and a fraction
of the energy is absorbed by an absorber plate. The air which is near the
absorber plate region is heated by greenhouse effect and therefore, the
density of air is lowered inside the setup and the chimney base. Hence,
the high dense outside air enters into the setup and there is a natural air
flow inside the setup. By natural convection mechanism, the heat
transfer takes place from hot ground surface to the adjacent air which
causes the temperature rise. Now the hot air moves up due to buoyancy
effect and enters the solar chimney where heat energy in air gets con-
verted to kinetic energy. The pressure difference between the chimney
base and the atmospheric air at the chimney outlet creates the air flow.
This pressure difference can be utilized to propel the solar wind turbine
which in turn rotates the generator for power generation.
In this study, a pilot solar chimney power plant consists of an air
collector with a diameter of 3.5 m and a chimney of diameter 0.6 m for
effective pressure drop across the turbine. The height of the chimney is
estimated as 6 m. Fig. 1 shows the basic outline diagram of a small scale
SUT. The main assumptions followed in this analysis are, (i) axisym-
metric air flow in the collector inlet, (ii) constant environmental con-
ditions including ambient temperature and inlet air temperature, (iii)
heat losses through the wall of the chimney are neglected, (iv) the air
follows the ideal gas law and (v) buoyancy force is considered in the
chimney.
2.1. The solar collector
The solar collector plays the important role like heat exchanger
which converts the incident solar radiation into useful stored energy in
the air. It comprises of two main elements namely glazed material
(admits radiation) and absorber plate. Table 1 gives the transmittance
453
Energy Conversion and Management 150 (2017) 451–462
of materials for solar collector cover.
Anti-refluxing glazed glass has been generally used to coat solar
collector cover since it can beam as much as 90% of the approaching
shortwave sun radiation while transmitting for all intents and purposes
none of the long wave radiation discharged outward by the absorber
plate. Glass with low iron substance has a moderately high transmit-
tance-absorbent product for sun radiation (roughly 0.85–0.91 at typical
rate), yet its transmittance is basically zero for the long wave warm
radiation (5.0–50 mm) discharged by heated surfaces. From Table 1,
the glazed glass has the highest transmittance value (0.85–0.9) com-
pared to the other materials. Copper was chosen as absorber plate
material because of its high thermal conductivity compared to steel and
aluminium.
2.2. Solar angles
With a specific end goal to compute the beam and diffuse radiation
on an inclined surface confronting due south, it is important to change
over the estimation of beam flux originating from the origin of the sun
to an equal esteem relating to the normal direction to the surface. If θ is
the angle between an incident beam of flux (Fbn) and the normal to a
plane surface, then the equivalent flux failing normal to the surface is
given by Fbn cos θ. The angle θ can be related by a general equation to
the latitude (L), the slope (ξ), the declination (δ), and the hour angle
(ω). The critical sun angles which are utilized as a part of calculations
of the direct and diffuse radiations are shown in Fig. 2. The collectors
ought to be situated specifically towards the equator, facing due south
in the northern hemisphere and north in the southern hemisphere. For
NIT Warangal, India the longitude is 79°35′ east and latitude (L)
= 18°00′ north.
The instantaneous hour angles are calculated during day time at
sixty minutes of intervals from sunset to sunrise at NIT Warangal, India.
These calculated hour angles are used to determine the angle of in-
cidence (θ) and solar zenith angle (Z) on each side of the collector. The
direct (or) beam radiation (Fbeam) and diffuse radiation (Fdiffuse) at each
interval are calculated from the angle of incidence (θ), zenith angle (Z),
tilt factors for both beam and diffuse radiation (rb and rd). From the
angle of incidence and the calculated beam radiation, the energy
transmitted through the glass cover is determined by using Snell’s law.
The net energy transmitted through the glass cover is the sum of the
transmitted beam radiation and the diffuse radiation.
2.3. Beam, diffuse and global radiation
The declination (δ) is the angle made by the line joining the centers
of the sun and the earth with the projection of this line on the equatorial
plane. The declination angle varies from a maximum value of +23.45°
on June 21 to a minimum value of −23.45° on December 21. It is zero
Fig. 1. Solar updraft tower (SUT) power plant. (All dimensions are in mm).
R. Balijepalli et al. Energy Conversion and Management 150 (2017) 451–462

Table 1 where ξ is the collector slope angle and γ = 0° for south facing collector
Solar collectors cover materials [7,25]. at the northern hemisphere.
ASHRAE [27] has given a method for estimating the hourly global
Material Transmittance
and diffuse radiation falling on horizontal surfaces under cloudless skies
Anti-refluxing glazed glass 0.85–0.9 (clear days). The equations are based on an exponential decay model in
Window glass 0.87 which the beam radiation decreases with increase in the distance tra-
Greenhouse glass 0.85
versed through the atmosphere. The global radiation FG reaching a
Polycarbonate 0.844
Fiber glass reinforced polyester 0.831 horizontal surface on the earth is given by Sukhatme and Nayak [26],
Corrugated fiber glass 0.78–0.79 FG = Fbeam + Fdiffuse (8)
where FG is hourly global radiation, Fbeam is hourly beam radiation and
on the two equinox days of March 21 and September 22. Fdiffuse is hourly diffuse radiation.
The solar declination angle (δ) [26] is estimated as, Now,
360(DY + 284) ⎤ Fbeam = Fbn cosZ (9)
δ = 23.45sin ⎡
⎣ 365 ⎦ (1) where Fbn is beam radiation in the direction of the rays and Z is angle of
where DY is day of the year. incidence on a horizontal surface, i.e. zenith angle.
The hour angle (ω) is an angular measure of time and is equivalent Thus,
to 15° per hour. It also varies from −180° to +180°. We adopt the FG = Fbn cosZ + Fdiffuse (10)
convention of measuring it from noon based on local apparent time
(LAT), being positive in the morning and negative in the afternoon. In the ASHRAE model [27], it is postulated that
The hour angle corresponding to sunrise or sunset (ωst) on an in-
Fbn = Ae⎣
⎡−( cosBZ ) ⎤⎦ (11)
clined surface facing due south (γ = 0°) [26],
ωst = cos−1 [−tanδtan(L−ξ)] and diffuse radiation,
(2)
Fdiffuse = CFbn (12)

Sunrise time: Tsunrise = ⎣

( ( ))
⎡ −ωst 60
15
+ 720⎤
⎦ where A, B and C are constants whose values have been determined on
60 (3) a month wise basis. These constants change during the year because of
seasonal changes in the dust and water vapor content of the atmosphere
⎡ ωst
Sunset time: Tsunset = ⎣
( ( ) ) + 720⎤⎦
60
15
and also because of the changing earth sun distance. The values of A, B
and C were initially given by Threlkeld and Jordan [27] and subse-
60 (4) quently revised by Iqbal [28].
Here, the time is mentioned as a fraction of an hour instead of hours The solar collector selected for this analysis is a sloped collector
and minutes. which is tilted at an angle to the horizontal. It therefore becomes ne-
The hour angle ω at any instant during day time is estimated [26], cessary to calculate the flux which falls on a tilted surface. Tilt factors
should be considered for beam and diffuse radiation calculations. The
360° (solar time−12) tilt factor for beam radiation is rb.
ω= ±
24 hours (5)
cosθ sinδ sin(L−ξ ) + cosδ cosωcos(L−ξ )
The hour angle ω is negative before solar noon and positive after rb = =
cosZ sinδ sinL + cosδ cosωcosL (13)
solar noon.
Solar incident angle (θ) on an inclined surface facing due south where Cos θ is beam radiation flux falling on a tilted surface and Cos Z is
(γ = 0°) [26], beam radiation flux falling on a horizontal surface.
The tilt factor rd for diffuse radiation is the ratio of the diffuse ra-
cosθ = sinδsin(L−ξ) + cosδcosωcos(L−ξ) (6) diation flux falling on the tilted surface to that falling on a horizontal
surface. The value of the tilt factor depends up on the distribution of
Solar zenith angle is estimated from,
diffuse radiation over the sky and on the portion of the sky dome seen
cos Z= sinδsin L+ cosδcosωcosL (7) by the tilted surface. Assuming that the sky is an isotropic source of

Fig. 2. Sun angles on collector surface.

454
R. Balijepalli et al. Energy Conversion and Management 150 (2017) 451–462

diffuse radiation, we have for a tilted surface with a slope (ξ), 3.1. Energy lost due to reflection on collector cover material

1 + cosξ
rd = When a beam of light travelling through a transparent medium 1
2 (14)
strikes the interface separating it from another transparent medium 2, it
Since 1 + cosξ is the radiation shape factor for a tilted surface with is reflected and refracted. The reflected beam has a reduced intensity
2
respect to the sky. and has a direction such that the angle of reflection is equal to the angle
Now the beam (or) direct radiation falls on tilted surface is calcu- of incidence. On the other hand, the directions of incident and refracted
lated from, beams are related to each other by Snell’s law which states that [26],
Angle of refraction,
B
Fbeam = A (cosθ) e⎡⎣−( cosZ )⎤⎦ (15)
⎡ sinθ ⎤
1⎥
Similarly the diffuse radiation falls on tilted surface is calculated θ2 = sin−1 ⎢
from,
⎢ n2 ⎥
⎣ n1 ⎦ ( ) (18)
1 + cosξ ⎤ where θ1 = Angle of incidence, n1 = refractive index of medium 1,
Fdiffuse = CFbn ⎡
⎢
⎣ 2 ⎥
⎦ (16) n2 = refractive index of medium 2.
The reflectivity of incident beam (ρ) is given by [26];
The global radiation FG reaching a sloped (or) tilted surface on the
ρ1 + ρ2
earth is given by [26] ρ=
2 (19)
⎡− B ⎤
( ) 1 + cosβ ⎤
FG = Fbeam + Fdiffuse = A (cosθ) e⎣ cosZ ⎦ + CFbn ⎡ where ρ1 and ρ2 being the reflectivity’s of the two components of po-
⎣ 2 ⎦ (17) larisation [26].
As per the above ASHRAE model [27], beam and diffuse radiations sin2 (θ −θ )
were calculated on 21st of every month of the year 2016, at NIT ρ1 = ⎛ 2 2 1 ⎞ ⎜ ⎟

⎝ sin (θ2 + θ1) ⎠ (20)

Warangal. Figs. 3 and 4 show that the comparison between the avail-
ability of solar radiation on 21st of every month of the year 2016. From tan2 (θ −θ )
both Figs. 3 and 4 the maximum solar global radiation was observed in ρ2 = ⎛ 2 2 1 ⎞ ⎜ ⎟

⎝ tan (θ2 + θ1) ⎠ (21)

summer (April/May). In the primary time frame; both daylight hour
and temperature are high. The second peak happens in August (harvest For beam radiation, the transmissivity of the collector cover can be
time) which is not so noticeable, because of a high temperature but obtained with adequate accuracy by considering reflection-refraction
rather short sun shining period (Fig. 4). In the winter (November/De- and absorption separately and is given by [26],
cember), again sufficient sun shining periods are available but the τ = τa τr (22)
temperature is low (Fig. 4). Subsequently, it brings about low global sun
radiation. These theoretically estimated global radiation values were where τr is transmissivity obtained by considering only reflection and
compared with India’s Meteorological department values, 2016 [29]. refraction and τa is transmissivity obtained by considering only ab-
The most extreme estimation of global radiation is 1059.38 W/m2 at sorption.
the time of April when the bright sunshine hour and normal tempera- 1
τr = (τr1 + τr 2)
ture are 9.17 h and 31.5 °C respectively [29]. The base estimation of 2 (23)
global radiation is 851.22 W/m2 at the time of December when the
where τr1 and τr2 being the transmissivity of the two components of
bright sunshine hour and normal temperature are 8.39 h and 21.6 °C
polarisation.
[29] respectively.
The estimated beam radiation values were compared with experi- 1−ρ1
τr1 =
mental data of same location (Warangal, India). Instantaneous beam 1 + ρ1 (24)
radiation was measured using a solar power meter (Tenmar TM 207).
1−ρ2
The experimental average value of beam radiation from February to τr1 =
June 2016 was 833 W/m2 [30] and the same by theoretical estimation 1 + ρ2 (25)
is 897.41 W/m2. The percentage difference noticed is 7.7%. The Transmissivity obtained by absorption [25],
average beam radiation throughout the year 2016 is estimated and it is
985 W/m2. The same estimated beam radiation for the year of 2016 is τa = e−Kδc /cosZ (26)
compared with the India Meteorological department [29] and the where K is glass extinction coefficient (m −1
) and δc is glass cover
maximum percentage variation is noticed as 6%.
Solar radiation parameters are estimated from Eqs. (1)–(17) and are 1200.00
listed in Tables 2 and 3. The parameters are estimated at noon time of
Global Radiation, W/m2

1000.00
21st of each month of 2016. Therefore, instantaneous hour angle at
solar noon (ω) is zero. The collector slope angle (ξ) and surface azi- 800.00 Jan
muthal angle (γ) are considered as 30° and 0° respectively [26]. Results
600.00 Feb
shows that direct radiation is higher in intensity to the global radiation
during the summer, post monsoonal and winter months due to clear sky Mar
400.00
conditions and variations in altitude angle. The average value of global Apr
radiation for 1st 6 month is 984.5 W/m2 and 2nd 6 month is 985.5 W/ 200.00 May
m2. June
0.00
06:30
07:30
08:30
09:30
10:30
11:30
12:00
12:30
13:30
14:30
15:30
16:30
17:30

3. Energy losses
Time, hours
In this section the energy lost in solar collector cover material due to
Fig. 3. Monthly average daily global radiation for first six months of 2016.
reflection and heat losses is estimated and discussed.

455
R. Balijepalli et al. Energy Conversion and Management 150 (2017) 451–462

1200.00 Transmissivity absorptivity product for diffuse radiation,

Global Radiation, W/m2

1000.00 τα
(τα )d =
[1−(1−α ) ρd ] (28)
800.00 July
Aug where α is absorptivity of absorber plate and ρd is diffuse reflectivity of
600.00
the collector cover,
Sep
400.00
Oct (ρd = (τa−τd ) (29)
200.00 Nov 2
The net global radiation absorbed per m area of the absorber plate
0.00 Dec [26] is;
FA . P = [(Fbeam)(τα) b] + [(Fdiffuse )(τα)d] (30)
Time, hours The total surface area of the collector cover is given by:
Fig. 4. Monthly average daily global radiation for last six months of 2016.
π
Ac = 4 ⎛ ⎞ θs (r /cos30)2
⎝ 360 ⎠ (31)
Table 2
Calculated parameters of solar beam, diffuse and global radiation for 1st six months of where r is radius of the collector plate and θs is central angle of sector
year 2016. (90°).
The theoretical radiant energy transmitted through collector cover,
Parameters Value
Eth = (Ac )(FG ) (32)
Day of the year, DY 21 52 70 111 141 172
Solar declination, δ −20.14° −11.23° −4.41° 11.58° 20.14° 23.45° The total incident flux absorbed by the absorber plate,
hour angle at sunrise 94.48° 90.83° 92.24° 97.8° 125.4° 130.3°
and sunset, ωst
EA . P = (AA . P )(FA . P ) (33)
Solar zenith angle, Z 38.12° 29.21° 22.4° 6.4° 2.15° 5.47°
Average sunshine hours available on 21st October 2016, (HS) = 7.3
Solar incident angle 8.12° 0.79° 7.6° 23.6° 32.15° 35.47°
at noon, θ The total theoretical energy absorbed by plate on 21st October
Constant, A (W/m2) 1202 1187 1164 1130 1106 1092 2016;
Constant, B 0.141 0.142 0.149 0.164 0.177 0.185
Constant, C 0.103 0.103 0.104 0.109 0.12 0.13
EA = EA . P HS (34)
Beam radiation, 790.45 790.45 916 952.1 925.8 902.7
Transmissivity estimations of solar collector cover are performed
Fbeam (W/m2)
Diffuse radiation, 96.55 96.55 100.75 107.27 112.4 115.9 using Eqs. (18)–(34) and are tabulated in Table 4.
Fdiffuse (W/m2)
Global radiation, Fg 887.01 887.01 1016.76 1059.4 1038.2 1018.6 3.2. Heat losses
(W/m2)

The radiative and convective heat transfer coefficients between the

thickness (m). The net transmissivity of the collector cover material is
given by Eq. (22). For diffuse radiation, the angle of incidence is taken
to be 60°. Therefore, angle of refraction can be estimated from Eq. (18),
the transmissivity of collector cover is estimated from Eq. (22).
Out of the fraction τ transmitted through the cover system, a part is
absorbed and a part reflected diffusively. Out of the reflected part, a
portion is transmitted through the collector cover and a portion is re-
flected back to the absorber plate. The process of absorption and re-
flection at the absorber plate surface goes on indefinitely, the quantities
involved being successively smaller. Transmissivity absorptivity pro-
duct (τα) indicates that the amount of flux absorbed in the absorber
plate with respect to the flux incident on the collector cover.
Transmissivity absorptivity product for beam radiation,
τα
(τα )b =
[1−(1−α ) ρd ]
collector cover and the absorber plate are determined to give the re-
sistance between the collector cover and the absorber plate. The line-
arized radiative exchange with the sky and wind related losses are
determined to evaluate the collector resistance. The collector resistance
and the resistance between the collector cover and absorber plate are
used to determine the top loss coefficient. From the top loss coefficient,
the rate of energy loss per m2 of collector base area is determined.
Useful energy is calculated as the difference between the energy
transmitted through the collector and the heat losses. The linearized
radiation coefficient is given by Maghrebiet al. [14]:
σ (TA + TC )(TA2 + TC2)
hL =
( 1
εA
+
1
εg )
−1
(35)
where σ is Stephan-Boltzmann constant, TA is absorber plate tempera-
(27) ture, Tc is collector cover temperature, εA is absorber plate emissivity, εg

Table 3
Calculated parameters of solar beam, diffuse and global radiation for last six months of year 2016.

Parameters Value

Day of the year, DY 202 233 264 294 325 355

Solar declination, δ −20.44° 11.75° −0.2° −11.75° −20.44° −23.45°
hour angle at sunrise and sunset, ωst 129.2° 125.9° 102.1° 90.3° 80.8° 77.2°
Solar zenith angle, Z 2.46° 6.23° 18.2° 29.75° 38.43° 5.47°
Solar incident angle at noon, θ 32.46° 23.77° 11.81° 0.26° 8.43° 35.47°
Constant, A (W/m2) 1202 1187 1164 1130 1106 1092
Constant, B 0.141 0.142 0.149 0.164 0.177 0.185
Constant, C 0.103 0.103 0.104 0.109 0.12 0.13
Beam radiation, Fbeam (W/m2) 906.5 916.35 907.2 828 925.8 725.7
Diffuse radiation, Fdiffuse (W/m2) 116.9 115.25 107.8 98.75 112.4 102.55
Global radiation, Fg (W/m2) 1023.3 1031.6 1015 926.74 1038.2 878.3

456
R. Balijepalli et al. Energy Conversion and Management 150 (2017) 451–462

Table 4 2 ⎡ TC −TSky ⎤
Transmissivity estimations of solar collector cover. hr = σ ∊g (TC + TSky )(TC2 + TSky )
⎣ TC−Ta ⎥
⎢ ⎦ (40)
Parameter Value Collector resistance is given by,
Hourly beam (or) direct radiation, (Fbeam) 827.99 W/m2 1
Hourly diffuse radiation, (Fdiffuse) 98.75 W/m2 R2 =
hr + hW (41)
Hourly global radiation, (FG) 926.74 W/m2
Refractive index of glass relative to the air, (n2/n1) 1.526 Therefore, the overall loss coefficient can be estimated from,
Glass extinction coefficient, (K) 15 m−1
Glass cover thickness, (δc) 5 mm 1
Ut =
Beam Diffuse R1 + R2 (42)
The rate of energy loss is estimated as,
Reflectivity of first component of polarisation, (ρ1) 0.043 0.185
Reflectivity of second component of polarisation, (ρ2) 0.043 0.001 QL = Ut Ac (TA−Ta) (43)
Angle of incidence, (θ1) 0.26° 60°
Angle of refraction, (θ2) 0.17° 34.58° Useful energy available,
Reflectivity, (ρ) 0.043 0.093
Transmissivity of first component of polarisation, 0.917 0.687 QU = EA . P−QL (44)
(τr1)
Transmissivity of second component of polarisation, 0.917 0.997 Energy lost per day can be calculated based on the rate of energy
(τr2) loss per hour of a day (7.3 h) and it is,
Transmissivity by considering reflection and 0.917 0.842
refraction, (τr)
Elost = HS QL. (45)
Transmissivity by considering absorption, (τa) 0.999 0.999
The useful energy available during sunshine hours in a day per m2 of
Transmissivity, (τ) 0.917 0.842
base area,
Absorptivity of copper plate (α) 0.94
Diffuse reflectivity of the cover system (ρd) 0.1579 E = EA−Elost (46)
Transmissivity absorptivity product for beam 0.870
radiation (τα)b The efficiency of the collector is,
Transmissivity absorptivity product for diffuse 0.799 QU
radiation (τα)d ηcoll =
Net global radiation absorbed by the absorber plate, 799.29 W/m2 Eth (47)
(FA.P)
The heat loss estimations are performed using Eqs. (35)–(47) and
Total surface area of the collector cover, (Ac) 12.831 m2
The theoretical radiant energy transmitted through 118,91 W are tabulated in Table 6. The efficiency of the collector for 21st October
collector cover (Eth) 2016 is estimated and it is 43.82% (Table 6). Similar collector effi-
The incident flux absorbed by absorber plate, (EA.P) 7690.05 W ciency is noticed in the study of Guo et al. [13]. In their numerical
Average sunshine hours available on 21-October- 7.3 h
work, the collector efficiency varied from 35 to 43%. Kalogirou [25]’s
2016 in Hyderabad. (nearest weather station to
NIT Warangal) [28]
collector efficiency is 45%. A solar tracking system was used in the
The total theoretical energy absorbed by plate on 21- 56137.36 Wh (or) study of Gitan et al. [12], therefore their collector efficiency increased
October-2016 (EA) 5834.88 Wh/m2 up to 51%. Maghrebi et al. [14] developed a sloped SUT plant and their
collector efficiency was 56%.

is collector cover emissivity. The collector absorber plate temperature is

4. Solar chimney (or) tower
considered as 80 °C and the ambient temperature is 30 °C.
The collector cover temperature is chosen as 35 °C (i.e. between the
It is somewhat difficult to find materials for solar chimney. The
ambient and the absorber plate temperatures). The convective heat
chimney material should be weather resistant, having light weight,
transfer coefficient hc [31] is given by,
strong in structure, withstand wind and overall loading conditions. The
thermal properties and availability of several materials were con-
s g ΔT ⎞
hc = (0.06−0.017) ⎛ ⎞ kL 3 ⎛ 2
⎜ ⎟ sidered, such as Nylon 66, polycarbonate pipes, polyethylene, poly-
⎝ 90 ⎠ ⎝ Aϑ ⎠
T (36) propylene, polyester, Poly tetra fluroethylene and polyvinylchloride
(PVC). Table 7 provides the list of possible chimney materials and their
where k is the thermal conductivity of air, L is the distance between the
properties. Among the various materials available, the polycarbonate
absorber plate and the collector, υ is kinematic viscosity of air, ΔT is
material is selected as chimney material and it can withstand air tem-
temperature difference between plate and air. All the input parameters
peratures of up to 150 °C.
needed for estimation of heat losses are mentioned in Table 5.
Resistance between the collector and absorber plate is,
Table 5
1 Input parameters of heat loss calculation.
R1 =
hL + hC (37)
Given data Value

Wind loss coefficient [26] is given by: Ambient air temperature, (Ta) 30 °C
Absorber plate temperature, (TA) 80 °C
h w = 5.7 + (3.8v) (38) Collector cover temperature, (Tc) 35 °C
Plate emissivity, (εA) 0.8
where v is average wind speed at 1 m height in NIT Warangal, India Collector emissivity, (εg) 0.9
conditions. Stephan-Boltzmann constant, (σ) 5.6 × 10−8 W/m2 K4
Sky temperature is, Thermal conductivity of air, (k) 0.026 W/m·K
Distance between plate and collector, (L) 0.06 m
3 kinematic viscosity of air, (υ) 1.5 × 10−5 m2/s
TSKY = (0.0552)(Ta) 2 (39) Temperature difference between absorber plate and 50 °C
ambient air, (ΔT)
The radiative heat transfer coefficient [26] referenced to the air Average wind speed, (v) 2.0 m/s
temperature can be estimated from,

457
R. Balijepalli et al. Energy Conversion and Management 150 (2017) 451–462

Table 6 Ech = 12Qa (52)

Estimated parameters of heat loss calculations.
Required collector base area is estimated based on the energy re-
Parameter Value quired for 12 h and it is,

Linearized radiation coefficient, (hL) 6.04 W/m2·K Ech

AR . C =
Convective heat transfer coefficient, (hc) 0.04 W/m2·K EA (53)
Wind loss coefficient, (hw) 15.2 W/m2·K
Radiative heat transfer coefficient, (hr) 21.32 W/m2·K Pressure difference between chimney base and surroundings
Resistance between the collector and absorber plate, (R1) 0.16 K/W [34,35] is estimated from,
Sky temperature, (Tsky) 18.14 °C
Collector resistance, (R2) 0.03 K/W 2
⎧ πFG ηc Rc ⎞ ⎛ gHch ⎞ ⎛ γ∞ Hch ⎞ ⎫
Overall heat loss coefficient, (Ut) 5.15 W/m2·K ΔP = 0.00353gHch ⎜⎛ ⎟−⎜ ⎟ +
⎨ ⎝ Cp ma ⎠ ⎝ 2Cp ⎠ ⎝ 2 ⎠ ⎬ (54)
Rate of energy loss, (QL) 2479.5 W ⎩ ⎭
Useful energy available, (QU) 5210.6 W
Efficiency of the collector, (ηcoll) 43.82% Volumetric flow rate of air through the chimney [9] is,
The total theoretical energy absorbed by plate on 21-October- 56137.36 Wh
Q = vmax Ach (55)
2016 (EA)
Energy lost through the collector during sunshine hours in a day, 18100.4 Wh The maximum theoretical power output of a turbine [9] is:
(Elost)
Useful energy (E) 38037.01 Wh Pt = (ΔP ) Q (56)
The variation of pressure drop across the turbine is shown in Fig. 5.
Table 7 Pressure drop is evaluated from Eq. (54) and it varies in between 0.602
Chimney materials and properties [13,32]. and 1.065 N/m2. The pressure drop value was experimentally evaluated
by Ahmed et al. [36] and it is 2.0 N/m2. Theoretical power output of
Thermosetting Tg (°C) Tm (°C) Thermal Density Specific
the system is in between 0.31 and 0.633 W. The power generation was
plastic material conductivity (kg/m3) heat (kJ/
(W/m·K) kg·K) experimentally evaluated by Ahmed et al. [36] and it is 0.1 W. Max-
imum power generation is noticed in the month of April due to high
Polyvinylchloride 87 212 0.14–0.28 1467 0.9 amount of global radiation.
(rigid PVC)
Chimney efficiency [8,37] is,
Nylon 6,6 100 265 0.24–0.3 1140 0.0017
Polycarbonate 150 265 0.19–0.22 1200 1.17 PTotal gHch
Polyester 73 265 – 1370 1.3–1.5 ηch = =
Qa Cp T (57)
Polyethylene −90 137 – 970 1.25
Polypropylene −14 176 – 905 1.92
The chimney efficiency is dependent on height and temperature.
Poly tetra −90 327 0.25 2200 1.172
fluroethylene Specific heat (Cp) is in the order of 1.005 kJ/kg·K and Th will always be
above 273 K (i.e. the denominator of the efficiency equation is in the
order of 273,000 or above). For the efficiency to be increased to a figure
With no wind turbine inside the chimney the maximum air velocity close to 100%, the chimney height must be in the order of 273,000/
[9] is, 9.81 = 278,28 m. This is three times the height of Mount Everest!
Increasing the chimney height to produce reasonable velocity is too
T
vmax = √ 2gHch ⎛1− a ⎞ ⎜ ⎟ expensive and will present major engineering challenges. The chimney
⎝ Th⎠ (48) efficiency varies with height at constant chimney temperature of 353 K.
where g is acceleration due to gravity. The analysis shows that although the chimney temperature is above the
The maximum velocity values of existing studies of Zhou et al. [4] ambient temperature, it should be kept as low as possible as Th appears
and Kasaeian et al. [32] are 3 m/s and 2.9 m/s respectively. In the on the denominator of the efficiency equation.
present study, it was estimated and varies between 1.82 and 2 m/s. It is The overall efficiency (ηo) of the plant [34] can be calculated from,
presented in Fig. 5. Pact
Area of the chimney is: ηo =
πFG Rc2 (58)
π
Ach 2
= ⎛ ⎞ DCh where Pact = Pt ηt , actual power output of the turbine, ηt is efficiency of
⎝4⎠ (49)
the turbine and Rc is collector radius.
The mass flow rate of air is: Maximum efficiency of the turbine (ηt) is considered as 66.67%
ma = ρa vmax Ach [10]. The overall efficiency of the plant throughout the year 2016 is
(50)
where ρa is density of air at chimney base. Velocity (m/s)
2.5
Heat flow rate is given by: Pres.drop (Pa)
Qa = ma Cp ΔT 2
(51)
Temperature of hot air is calculated from Eq. (44) and (51) and it is 1.5
shown in Fig. 6. where Cp is specific heat of air, ΔT is temperature
difference of air inside and outside collector. This temperature differ- 1
ence is estimated and it varies in between 9 and 15 °C and it is men-
tioned in Table 8. This value is quite comparable with Pasumarthi and 0.5
Sherif [33]’s value of 13.9 °C. Haaf et al. [2] and Haaf [3]’s pilot plant
in Manzanares, Spain gave 17.5 °C temperature difference. Zhou et al. 0
Jan Feb Mar Apr May June July Aug Sep Oct Nov Dec
[4] and Gitan et al. [12]’s system developed 15 and 17 °C respectively. 21st 21st 21st 21st 21st 21st 21st 21st 21st 21st 21st 21st
Depends upon the geographical location the temperature difference was Day of the Year 2016
varied from 11 to 28 °C in different studies [4,5,7,21,22,33].
Fig. 5. Variation of velocity, pressure drop and power output in the year 2016.
Therefore energy required for 12 h operation is,

458
R. Balijepalli et al. Energy Conversion and Management 150 (2017) 451–462

60 5.2. Sensible heat storage materials

Hot air temperature, C

50
Thermal energy is accumulated in the storage media while there is
40 an increase (or) decrease of temperatures in sensible heat storage pro-
30
cesses. Table 9 gives the important properties of solid state thermal
energy storage materials. These materials work between the tempera-
20 ture range of 200–1200 °C. Materials such as fire bricks, sand-rock
minerals and concrete have superior thermal conductivities varying
10
from 1.0 to 7 W/m·K [24,37]. The materials given in Table 9 are less
0 price ranging from 3.35 to 334 Rupees/kg (0.05–5.00 US$/kg). The
Jan Feb Mar Apr May June July Aug Sep Oct Nov Dec main advantage of a sensible thermal energy storage is in low price
21st 21st 21st 21st 21st 21st 21st 21st 21st 21st 21st 21st
when compared with the price of latent heat storage materials which
Day of the Year 2016 normally ranges from 286 to 22,310 Rupees/kg (4.186–326 US$/kg).
Fig. 6. Variation of hot air (Th) temperature in the year 2016. Among the above mentioned solid state thermal energy storage
materials, the sand rock minerals (1.3 kJ/kg·K), magnesia fire bricks
(1.15 kJ/kg·K) and soil gravelly (1.84 kJ/kg·K) have large heat capa-
Table 8
Input and estimated parameters of chimney.
cities. Cast iron and steel have the highest thermal conductivity
(37–40 W/m·K) when compared to other storage materials. Soil and
Parameter Value sand rock minerals have better thermo physical characteristics at a
lower cost. Therefore these materials are selected as thermal energy
Absorber plate temperature, (TA) 80 °C
Ambient or air temperature, (Ta) 30 °C
storage materials for making solar chimney power plant. Thermal en-
temperature difference of absorber plate (TA) and hot air (Th) 40 °C ergy can be stored cheaply in rocks, where the stored energy is given
Diameter of chimney, (Dch) 0.6 m by,
Height of the chimney, (Hch) 6m
Density of air at chimney base 1.005 kg/m3 Et = mCm ΔT (When using the mass specific heat) or (59)
Maximum air velocity, (vmax) 1.94 m/s
Area of the chimney, (Ach) 0.2827 m2 Et = VCV ΔT (When using the volumetric specific heat) (60)
Mass flow rate of air, (ma) 0.55 kg/s
Air heat transfer rate, (Qa) 5.56 kW
Consider temperature difference as 15 °C.
Energy required for 12 h operation (Ech) 66.72 kWh
Required collector base area (AR.C) 8.92 m2
The required energy,
Pressure drop inside the chimney, (ΔP) 0.55 N/m2
Et = 12QU (61)
Maximum theoretical power output of a turbine (Pt) 0.37 W
Efficiency of the turbine (ηt) [9] 66.6%
where Cm and Cv are mass and volumetric specific heat of rock re-
Actual power output of a turbine (Pact) 0.25 W
Efficiency of the chimney (ηch) 0.0193% spectively.
Overall efficiency of the plant (ηo) 0.0015% Total mass of the rocks required,
Et
m=
estimated using Eq. (58) and it is presented in Fig. 7. It is in the range Cm ΔT (62)
between 0.0017 and 0.0028% and this value range is almost similar of
Total volume of the rocks required:
Pasumarthi and Sherif [33]’s value of 0.0013–0.0023%. The maximum
overall efficiency (0.0028%) is noticed in the month of April 2016. All Et
V=
the input and estimated parameters of chimney are mentioned in CV ΔT (63)
Table 8.
The maximum thermal energy required for producing the power of
0.633 W was estimated as 80.64 kWh and it is happened in April 2016.
5. Energy storage materials The stored energy is used for continuous operation during the nights
and also for periods of overcast weather conditions.
In this section, the thermal energy storage materials are discussed These rocks are placed in the space below the collector surface. The
and appropriate materials are selected. volume below the absorber plate is given by,

Vav = Ab hg (64)

5.1. Thermal energy storage where Ab is base area

There are three significant features that are essentially considered 0.0030
Overall Efficiency, %

for making the design of a solar thermal energy storage system: tech- 0.0025
nical characteristics, price effectiveness and effect on environment. An
important element when considering the technical feasibility of a solar 0.0020

thermal energy storage system is a system that shows superior technical 0.0015
properties. The thermal storage capacity should be on the higher side in
0.0010
order to reduce the total volume occupied by the system and to improve
the efficiency of the system. Secondly, the rate of heat transfer between 0.0005
the storage material and transfer fluid (air) must be high for the release
0.0000
(or) absorb the thermal energy at uniform speed. Thirdly, the storage Jan Feb Mar Apr May June July Aug Sep Oct Nov Dec
material should possess excellent stability to prevent mechanical and 21st 21st 21st 21st 21st 21st 21st 21st 21st 21st 21st 21st
chemical deterioration after a finite number of periodic thermal cycles. Day of the Year 2016
Fig. 7. Variation of overall efficiency in the year 2016.

459
R. Balijepalli et al. Energy Conversion and Management 150 (2017) 451–462

Table 9 mentioned in Table 11. Wood and bamboo turbine blade materials are
Solid state sensible heat storage materials [7,38]. lesser cost materials for turbine blade. Wood was used in ancient days
because of its greater availability and elastic limit. But it has not been
Storage Working Density Thermal Specific
material temperature (°C) (kg/m3) conductivity (W/ heat (kJ/ encouraged in recent decades because of its poor moisture resistance
m·K) kg·K) and low stiffness.

Sand-rock 200–300 1700 1.0 1.30

minerals
6.2. Turbine blade profile design
Reinforced 200–400 2200 1.5 0.85
concrete
Cast iron 200–400 7200 37.0 0.56 Turbine blades are designed in such a way as to turn the low density
NaCl 200–500 2160 7.0 0.85 air flow through a chosen angle with minimum friction losses. The
Cast steel 200–700 7800 40.0 0.60 principal load acting on the turbine blades is the bending force due to
Silica fire 200–700 1820 1.5 1.00
upward force, which causes maximum bending stress at the root of the
bricks
Magnesia fire 200–1200 3000 5.0 1.15 turbine blade. Generally three categories of blade profiles are used in
bricks turbines [19]. They are,
Dry bricks 200–800 1800 0.5 0.84
Soil (gravelly) – 2040 0.59 1.84
• NACA 4 digit series
• NACA 5 digit series
Soil (clay) – 1450 1.28 0.88

• NACA 6 digit series.

π
Ab = ⎛ ⎞ DA2 The wind turbine chosen for this study is a lift type HAWT
⎝4⎠ (65)
(Horizontal axis wind turbine) for the reason that lift type wind tur-
DA is diameter of absorber plate = 3.5 m. bines can possibly deliver more power than drag type turbines [39].
The height (hg) of the ground is selected as 100 mm for providing HAWT does not require a starting motor to get up to operating speed; it
the heat storage materials. The available volume is therefore, requires that the area projected by the blades is facing perpendicular to
Va = 0.962 m3. The input and process parameters of thermal energy the direction of the wind.
storage system are estimated and tabulated in Table 10. A NACA0012 blade design has been chosen for this work because it
allows ease of manufacture due to its flat high pressure side. The blade
6. Solar wind turbine thickness for a NACA0012 blade is given as 12% of the chord. The
velocities and angles are shown in Fig. 8.
The turbine blade material selection and design part are explained The maximum angular speed of the shaft considered in this work is
in this section. 500 rpm. The inside diameter of the chimney is 0.6 m; therefore the
diameter of the blade is 0.57 m after giving a clearance length of
6.1. Turbine blade selection 0.03 m. These types of blades and turbine setup are available in the
local markets in India. The tip speed ratio for a rotor is given by,
There are various turbine blade materials selected and their prop-
ωR
erties analyzed (Table 11). λ=
v1 (66)
The rotor blades have to endure very large deflections. The stiffness
should be much and it must be moisture resistant. Steel blades and where R = Turbine blade tip radius = 0.285 m, ω = rotational speed
other steel alloys are prominently used by industries for their higher 2πN
of the turbine in rpm, ω = 60 radians/s .N = number of rotations of
inertia which enhances the rotation speed and power output. Later, shaft Angle between the relative wind and plane of rotation is,
steel also was eliminated from the list of turbine blade material because
of its huge weight and low fatigue [22]. 2 R
∅ = ⎛ ⎞ tan−1 ⎛ ⎞
Aluminium seemed to be the next preferred blade material which ⎝3⎠ ⎝ λr ⎠ (67)
has one - third the density of steel and is therefore a light weight metal.
Aluminium has less fatigue level than steel, but it is ductile and a good Optimal turbine blade pitch angle is,
heat conductor. The other advantages of Aluminium are, it has low
tensile strength and it has a lower price than other higher conductive 2 R
β = ⎡ ⎛ ⎞ tan−1 ⎛ ⎞ ⎤−α
metals and good reliability. ⎣⎝ 3 ⎠ ⎝ λr ⎠ ⎦ (68)
The composites which are a combination of vinyl esters, epoxies and
polyesters have great thermal, mechanical and chemical properties. The where α is angle of attack on to the blade and r is radius of blade
glass fibers which are also used for turbine blades are slightly foggy segment from the hub.
with isotropic characteristics. These glass fibers are made of electrical Optimal chord length as a function of blade radius ‘r’ is,
glass (E-glass) and have good heat resistant property [38]. Also such
strengthened glass materials have greater mechanical and electrical Table 10
Input and process parameters of thermal energy storage system.
properties. E-glass can be obtained from fortifying mat, woven texture
and meandering. These materials have enough stiffness and high tensile Parameter Value
strength for making turbine blades. Carbon fibers have also been used
as turbine blade material in recent years as they have high stiffness and Mass specific heat of rock, (Cm) 1.3 kJ/kg·K
Volume specific heat of rock, (cV) 2.9 kJ/m3·K
tensile strength. These materials have low density, low weight and Diameter of energy storage box 3.5 m
therefore are a suitable option for the application of low turbine speed Height of energy storage box 100 mm
application. From Table 11, it can be seen that the properties of glass Base area, (Ab) 9.621m2
and carbon fiber composite are higher than that of other composites. Required thermal energy, (Et) 136.93 MJ
Mass of the rocks required for releasing 1 kWh energy, (mr) 185 kg
Therefore, the blades of large-scale as well as small scale wind turbines
Volume of the rocks required for releasing 1 kWh energy, (Vr) 82.76 m3
are mainly made of fiber-reinforced composite including glass and Available volume below absorber plate (Va) 0.9621 m3
carbon fibers. Price analysis is also made for various materials and

460
R. Balijepalli et al. Energy Conversion and Management 150 (2017) 451–462

Table 11 for other positions along the blade from root to tip.
Turbine blade materials and properties [9,19,22]. Effect of the tip speed ratio, blade pitch angle and relative wind
angle at different blade segments are plotted in Figs. 9–11. In the case of
Turbine blade Thermo – mechanical properties Price
materials per kg air velocity 2 m/s, there is not sufficient lift force to rotate the turbine
Strength Stiffness Energy/ Energy/ ($) blades. Maximum power generation under this condition is 0.012 W.
(MPa) (GPa) stiffness (J/ strength Optimal chord length is achieved at second segment of the blade. Drop
Nm) (kJ/Nm) of the relative wind angle and blade pitch angle between two successive
Aluminium 300 70 11.4 2.67 3.1 blade elements is high at lower wind speeds. Higher tip speed ratio is
Steel (Gr.43) 275 210 2.4 1.87 3.88 possible at lower wind speeds and it demands reduced chord widths
GFRP (UD 300 40 6.3 0.83 2.4 which leads to narrow blade profiles. This follows to minimum material
Glass) usage and lower production costs. However centrifugal and aero-
CFRP (UD 900 125 4.0 0.56 2.71
Carbon)
dynamic forces are enhanced with higher tip speeds. The increased
Wood 120 16 0.24 0.032 0.54 forces showed that very difficult to preserve structural integrity and
Bamboo 180 25 0.15 0.021 0.16 aerodynamics of blade design which leads to failure of blade. Noise is
also be enhanced along with the higher tip speed ratios. A blade which
is basically designed for higher relative wind speeds produces less
torque at lower wind speeds. High tip speed ratio produces a rotor blade
with a low solidity ratio.

7. Economic analysis on SUT materials

Some of the polymer pipes (chimney material) cannot withstand a

temperature greater than 70 °C. If we consider the low cost material
such as window and Plexi glass for collector cover, there may be a
chance of damage at heavy air gust loads. Stability of the system is also
very important in the selection of materials. Nowadays Glass fiber re-
inforced plastic (GFRP) material is used for making rotor blades be-
cause of less weight, greater stability and withstand to various static
and dynamic forces when compared to all other materials. Density of
the GFRP material is very less compared to steel, aluminium, wood, etc.
Gravel or rocks which are freely available can be used to store the solar
energy when the solar energy exceeds the required level during day
time.

8. Conclusions
Fig. 8. Velocities and angles.
A modest attempt was made to estimate the performance para-
1 16πr ⎞ 2 ⎡ 1 R meters of an SUT with the chimney and solar collector plate diameter of
COPT = ⎛ ⎞ ⎛ ⎜ sin ⎛ ⎞ tan−1 ⎛ ⎞ ⎤
⎟
0.6 m and 3.5 m respectively. The average global radiation of year 2016
⎝ B ⎠ ⎝ CL ⎠ ⎣⎝ 3 ⎠ ⎝ λr ⎠ ⎦ (69)
was estimated as 985 W/m2 and it is compared with experimental data
Solidity ratio is of India Meteorological department. Inside the chimney the maximum
cN air velocity was found as 2 m/s. The maximum theoretical power
σs = output of SUT was 0.633 W. As per the design of this present system,
2πr (70)
the maximum overall efficiency of the plant was estimated as 0.0028%.
Lift force can be calculated by If we need to achieve 100% efficiency of the plant, then the chimney
1 height should be 27 km, which is impossible for all practical purposes.
FL = CL ρw 2 (bc ) The heat losses, pressure loss, solar radiation and energy storage esti-
2 (71)
mations were performed. Maximum pressure drop inside the chimney
Drag force can be calculated by
1 8.00
FD = CD ρw 2 (bc ) Tip speed ratio
2 (72) 7.00 Optimal chord length (m)
where CL is coefficient of lift, CD is coefficient of lift, ρa is density of air 6.00 Solidity ratio
(kg/m3), w is relative wind speed (m/s), b is width of the blade section
5.00
(m), c is length of the chord line (m) and B is number of blades.
For efficient blade design, the number of rotor blades considered 4.00
three, angle of attack is 7° and coefficient of lift is 0.9. Wind turbine 3.00
parameters such as relative wind angle, pitch angle, airfoil and chord
length are estimated by blade element momentum (BEM) theory at 2.00
different turbine blade segments and details are shown in Figs. 9–11. 1.00
From Fig. 9 it is noticed that the optimal chord length of the blade
0.00
(0.284 m) is achieved at the relative radius (r/R) of 0.3. 0 0.05 0.1 0.15 0.2 0.25 0.3
The various angles and design parameters such as tip speed ratio Blade segment Radius (m)
(λ), relative wind angle (ϕ), blade pitch angle (β), relative chord length
(c/R) and solidity ratio (σ) are calculated for first blade segment having Fig. 9. Variation of tip speed ratio (λ), optimal chord length (Copt) and solidity ratio (σ)
with different blade segment radius.
a radius (r) of 0.04275 m from the hub. These calculations are repeated

461
R. Balijepalli et al. Energy Conversion and Management 150 (2017) 451–462

0.140 [6] Shirvan KM, Mirzakhanlari S, Mamourian M, Abu-Hamdeh N. Numerical in-

Power (W) vestigation and sensitivity analysis of effective parameters to obtain potential
0.120 maximum power output: a case study on Zanjan prototype solar chimney power
Lift Force (N)
plant. Energy Convers Manage 2017;136:350–60.
0.100 Drag Force (N) [7] Tian Y, Zhao CY. A review of solar collectors and thermal energy storage in solar
0.080 thermal applications. Appl Energy 2013;104:538–53.
[8] Schlaich J, Bergermann R, Schiel W, Weinrebe G. Design of commercial solar up-
0.060 draft tower systems – utilization of solar induced convective flows for power gen-
eration. J Sol Energy Eng, Trans ASME 2005;127:117–24.
0.040 [9] von Backström TW, Gannon AJ. Solar chimney turbine performance. J Sol Energy
Eng, Trans ASME 2003;125:101–6.
0.020 [10] Durrani N, Hameed H, Rahman H, Chaudhry SR. A detailed aerodynamic design
and analysis of a 2D vertical axis wind turbine using sliding mesh in CFD. In: 49th
0.000
AIAA aerospace sciences meeting including the new horizons forum and aerospace
0.00 2.00 4.00 6.00 8.00
exposition 4–7 Orlando, Florida; 2011.
Tip Speed Ratio
[11] Gitan AA, Abdulmalek SH, Dihrab SS. Tracking collector consideration of tilted
Fig. 10. Variation of power, lift and drag force with tip speed ratio. collector solar updraft tower power plant under Malaysia climate conditions.
Energy 2015;93(2):1467–77.
[12] Guo P, Li J, Wang Y, Wang Y. Numerical study on the performance of a solar
60.0 Relative wind angle (°)
chimney power plant. Energy Convers Manage 2015;105:197–205.
[13] Hu S, Leung DYC, Chan JCY. Impact of the geometry of divergent chimneys on the
50.0 Blade pitch angle (°) power output of a solar chimney power plant. Energy 2017;120:1–11.
[14] Maghrebi MJ, Nejad RM, Masoudi S. Performance analysis of sloped solar chimney
power plants in the southwestern region of Iran. Int J Ambient Energy 2016:1–8.
40.0
[15] Ozgen F, Esen M, Esen H. Experimental investigation of thermal performance of a
double flow solar air heater having aluminium cans. Renewable Energy
30.0 2009;34:2391–8.
[16] Esen H, Ozgen F, Esen M, Sengur A. Artificial neural network and wavelet neural
20.0 network approaches for modeling of a solar air heater. Expert Syst Appl
2009;36:11240–8.
10.0 [17] Esen H, Ozgen F, Esen M, Sengur A. Modelling of a new solar air heater through
least squares support vector machines. Expert Syst Appl 2009;36:10673–82.
0.0 [18] Esen M, Yuksel T. Experimental evaluation of using various renewable energy
0 0.05 0.1 0.15 0.2 0.25 0.3 sources for heating a green house. Energy Build 2013;65:340–51.
[19] Peter JS, Richard JC. Wind turbine blade design. Energies 2012;5:3425–49. ISSN
-10.0
Blade Segment Radius (m) 1996-1073.
[20] Asress MB, Aleksander S, Dragan K, Slobodan S. Numerical and analytical in-
Fig. 11. Variation of pitch and wind angles with different blade segment radius. vestigation of vertical axis wind turbine. FME Trans 2013;41:49–58.
[21] Soren G. A text book of wind turbines, 2nd ed. University College of Arhaus; 2009.
[22] Xua J, Zhang W, Chunxia W. Study on material selection and manufacturing process
was 1.065 N/m2. The energy required for producing the power of of wind turbine blades. J Adv Mater Res 2011;150–151:1621–4.
0.633 W was estimated as 80.64 kWh. Rocks were considered as energy [23] Tingzhen M, Wei L, Guoliang X. Analytical and numerical investigation of the solar
chimney power plant systems. Int J Energy Res 2006;30(11):861–73.
storage material in this work. The amount of rocks needed for releasing [24] Harte R, Höffer R, Krätzig WB, Mark P, Niemann HJ. Solar updraft power plants:
1 kWh energy was estimated as 184.6 kg or 82.75 m3. Appropriate engineering structures for sustainable energy generation. Eng Struct
materials were selected for turbine, chimney and solar collector after 2013;56:1698–706.
[25] Kalogirou SA. Solar thermal collectors and applications. Prog Energy Combust Sci
analysing their material characteristics and properties. Design calcula- 2004;30(3):231–95.
tions were made and an optimized chord length was selected for the [26] Sukhatme SP, Nayak JK. Solar energy-principles of thermal collection and storage.
turbine blade. All the input and estimated parameters were tabulated so 3rd ed. New Delhi (India): Tata McGraw Hill Education Private Limited; 2008.
[27] Threlkeld JL, Jordan RC. Direct solar radiation available on clear days. ASHRAE
that anyone could make their own prototype or real model of solar
Trans 1958;64:45.
updraft tower (SUT) plant after scaling this data. [28] Iqbal M. An introduction to solar radiation, 1st ed. Canada: Academic Press; 1983.
[29] Tyagi AP. Solar radiant energy over India. Meteorological Department, Ministry of
Earth Sciences, Government of India; 2016.
Acknowledgement
[30] Abhay L, Chandramohan VP, Raju VRK. Design, development and performance of
indirect type solar dryer for banana drying. Energy Proc 2017;109:409–16.
The authors acknowledge the financial support provided by (i) [31] Coetzee H. Design of solar chimney to generate electricity employing a convergent
Science & Engineering Research Board (SERB), Department of Science nozzle. Ph.D thesis. Private Bag 0082, Gaborone (Botswana, South Africa):
Botswana Technology Center; 1999.
and Technology (DST), New Delhi - 110 070, India, Grant No. File [32] Kasaeian AB, Heidari E, Vatan SN. Experimental investigation of climatic effects on
Number: EEQ/2016/000111 and (ii) Center of Excellence (CoE) under the efficiency of a solar chimney pilot power plant. Renew Sustain Energy Rev
TEQIP – II, National Institute of Technology Warangal, Warangal, India, 2011;15:5202–6.
[33] Pasumarthi N, Sherif SA. Experimental and theoretical performance of a demon-
Ref. No.: TEQIP/CoE/2016. stration solar chimney model—Part II: experimental and theoretical results and
Authors acknowledge the support received by way of proof reading economic analysis. Int J Energy Res 1998;22:443–61.
from Dr. M.R. Vishwanathan, Assistant Professor of English, Humanities [34] Zhou X, Yang J, Xiao B, Hou G, Xing F. Analysis of chimney height for solar chimney
power plant. Appl Therm Eng 2009;29(1):178–85.
and Social Science Department, NIT Warangal, India. [35] Asnaghi A, Ladjevardi SM. Solar chimney power plant performance in Iran. Renew
Sustain Energy Rev 2012;16:3383–90.
References [36] Ahmed A, Driss Z, Abdallah B, Abid MdS. Experimental and numerical study of the
impact of the collector roof inclination on the performance of a solar chimney
power plant. Energy Build 2017;139:263–76.
[1] Wengenmayr R, Bührke T. Renewable energy: sustainable energy concepts for the [37] Larbi S, Bouhdjar A, Meliani K, Taghourt A, Semai H. Solar chimney power plant
energy change, 2nd ed. Germany; 2013. with heat storage system performance analysis in South Region of Algeria. In: 2015
[2] Haaf W, Freidrich K, Mayr G, Schlaich J. Solar chimneys Part 1: principle and 3rd International renewable and sustainable energy conference (IRSEC). IEEE
construction of the pilot plant in Manzanares. Int J Sol Energy 1983;2:3–20. Conference Publications. p. 1–6. http://dx.doi.org/10.1109/IRSEC.2015.7454948.
[3] Haaf W. Solar chimneys Part 2: preliminary test results from the Manzanares pilot [38] Kaygusuz K. The viability of thermal energy storage. Energy Sources
plant. Int J Sol Energy 1984;2:141–61. 2010;21(8):745–55.
[4] Zhou XP, Wang F, Ochieng RM. A review on solar chimney power technology. [39] Ebaid MSY, Al-Hamdan QY. Optimization techniques for designing an inward flow
Renew Sustain Energy Rev 2010;14:2315–38. radial turbine rotor. Proc Inst Mech Eng, Part A: J Power Energy
[5] Pretorius JP, Kröger DG. Critical evaluation of solar chimney power plant perfor- 2004;218(8):655–68.
mance. Sol Energy 2006;80:535–44.

462