hassan2018

Accepted Manuscript
Numerical investigation on performance of solar chimney power plant by varying
collector slope and chimney diverging angle
Aakash Hassan, Majid Ali, Adeel Waqas
PII:
S0360-5442(17)31716-4
DOI:
10.1016/j.energy.2017.10.047
Reference:
EGY 11691
To appear in:
Energy
Received Date: 7 November 2016
Revised Date:
8 October 2017
Accepted Date: 10 October 2017
Please cite this article as: Hassan A, Ali M, Waqas A, Numerical investigation on performance of solar
chimney power plant by varying collector slope and chimney diverging angle, Energy (2017), doi:
10.1016/j.energy.2017.10.047.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
2
Numerical investigation on performance of solar chimney power
plant by varying collector slope and chimney diverging angle
3
Aakash Hassan∗, Majid Ali, Adeel Waqas
4
5
6
7
8
9
U.S. Pakistan Center for Advanced Studies in Energy (USPCAS-E),
National University of Sciences and Technology (NUST), H-12 Sector Islamabad, Pakistan
RI
PT
1
Email Addresses: [email protected]|[email protected],
[email protected]|[email protected]
Abstract
In this study, a parametric three dimensional computational fluid dyamics (CFD) analysis of
solar chimney power plant was performed to illustrate the effects of collector’s slope and
chimney diverging angle on performance of Manzanares prototype. Numerical models were
incorporated with solar load, DO (discrete ordinates) and RNG k-ɛ turbulence models. Firstly,
CFD simulations results were validated by comparing them with the experimental data of
Manzanares prototype and then, on the basis of robustness of numerical methods adopted,
several numerical simulations were performed on varying collector’s slope and chimney
diverging angles to improve the performance of solar chimney. Numerical simulations were
performed at chimney diverging angle = 10-30 and at collector slope = 40, 60, 80 and 100
sequentially, while all other parameters were kept constant. Based on computed results, it was
discovered that both velocity and temperature increases with increasing collector’s slope due to
enhanced heat transfer and mass flow rate, but simultaneously higher collector slopes also
deteriorate the smooth air flow by developing vortices and recirculation of air, which obstructs
the air flow and may reduce the overall performance. In addition, chimney diverging angle =10
raises the velocity from 9.1 m/s to a remarkable value of 11.6 m/s; therefore, this diverging
chimney approach is conceived to be a beneficial tool in improving performance of solar
chimney power plant.
27
Keywords
28
Solar chimney power plant; Computational fluid dynamics; Collector slope; Diverging chimney
29
1. Introduction
30
31
32
33
34
35
Needless to say, energy plays a role of backbone for the socio-economic development of any
country. Incessantly rising population and high living standards ask for endless energy resources
to fulfill heightening energy demand. It is pragmatic that non-renewable energy resources (fossil
fuels) are limited in nature and are being over utilized, so need to be restrained. Likewise, farreaching consumption of fossil fuels has polluted our environment severely, endangering life on
our planet. This entire scenario necessitates the advancements in clean and sustainable energy
AC
C
EP
TE
D
M
AN
U
SC
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
∗ Correspondence information: Aakash Hassan, U.S. Pakistan Center for Advanced Studies in Energy (USPCAS-E),
National University of Sciences and Technology (NUST) H-12 Sector Islamabad, Pakistan, +923002868138
1
ACCEPTED MANUSCRIPT
production. Solar chimney power plant or solar updraft tower is a renewable type power
generation technology, having a promising potential for large scale usage of solar energy. Its
working principle is based upon three technologies named as, the chimney, collector and wind
turbine. Solar radiations fall over the transparent collector heating the cold air beneath it, as hot
air rises, chimney raised at the center of collector provides air an upthrust. This kinetic energy
gained by air is used to run wind turbine generating electricity. This basic working principle of
SCPP is illustrated in Fig.1.
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
The basic idea of producing electricity by combination of greenhouse effect and chimney was
intimated by J. Schlaich. Later on, in 1983 J. Schlaich along with his team designed and
constructed the first working solar chimney pilot plant in Manzanares, Spain. This prototype
operated successfully for almost 8 years producing peak output of 50kW and constituted
collector’s diameter of 240m, chimney’s height of 200m and a chimney’s diameter of 10m [1, 2].
Afterwards, the results obtained from Manzanares plant and fundamental governing equations
for predicting the performance of SCPP were also reported. This unique idea of power
generation attracted many scientists; consequently a lot of efforts were made to explore this
technology. Y.J. Dai et al. evaluated the performance of solar chimney in Northwestern regions
of China by developing a simplified mathematical model and proposed a solar chimney that
could produce 110-190 kW monthly, supplying nearby villages in Northwestern regions of China
[3]. K.S. Ong and C.C. Chow predicted the performance of solar chimney by solving steady state
heat transfer equations using a thermal resistance network. The effect of different air gaps on the
output were studied and mathematical model results were verified by setting up an experimental
model [4]. Some investigators have also reported valuable comprehensive analytical models for
predicting the performance of solar chimney power plant by taking into account the factors like
friction loss, turbine loss and quality of soil, collector etc. [5–8].
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
Accompanying the advancement in computer technology, the use of CFD softwares in studying
the fluid flow and heat transfer has increased tremendously during the last decade due to its
powerful visualization capabilities. Roozbeh Sangi et al. performed numerical simulations by
iterative technique and predicted the performance of solar chimney through two dimensional
axisymmetric modeling in CFD software FLUENT. The results of two approaches were
compared ensuring good agreement [9]. Ehsan Gholamalizadeh and Man-Hoe Kim included the
effects of heat transfer by radiations in the solar chimney by using two band radiation model and
concluded that considering greenhouse effect is important for predicting the performance of solar
chimney in an accurate manner [10]. The effects of geometric parameters on performance of
small scale solar chimney were studied by Sandeep K. Patel et al. in ANSYS CFX. Results at
different collector’s inlet and outlet diameters, chimney diameters and divergence angles were
examined and the best configuration for solar chimney was suggested [11]. Ming Tingzhen
performed numerical simulations on Spanish prototype solar chimney coupled with a three-blade
turbine. The results indicated output power to be slightly greater than the experimental value;
they also designed a MW graded solar chimney power plant [12]. Ehsan Shabahang Nia et al.
numerically investigated the effects of three different passive flow control devices on heat
AC
C
EP
TE
D
M
AN
U
SC
RI
PT
36
37
38
39
40
41
42
2
ACCEPTED MANUSCRIPT
transfer and fluid flow inside solar chimney. The investigations revealed that flow control
devices increase the heat transfer and velocity magnitude in solar chimney due to vortices
developed by obstacles, resulting in better mixing of hot and cold air [13]. Penghua et al.
conducted a three dimensional CFD analysis of solar chimney with a real turbine. It was found
that fan model is also a convenient approach to predict the turbine performance in solar chimney;
additionally the influence of sun zenith angle and turbine performance with rotational speed were
also studied, and were found to be necessary in accurately predicting the performance of solar
chimney [14]. Amir Rezaei Shahreza et al. developed a novel solar chimney which utilizes
intensifiers to enhance the heat flux radiated by sun [15]. Recently, Ehsan Gholamalizadeh and
Man Hoe Kim performed a CFD analysis of solar chimney with an inclined collector by
increasing the distance between ground and chimney base concluding that, inclined collector
enhances the mass flow rate in system [16].
88
89
90
91
92
93
94
95
96
97
98
Although worthful contributions were made to analyze performance and develop more insight of
SCPP but, most of them studied similar parameters of SCPP such as: height of chimney, turbine
pressure drop, diameter of collector and heat storage medium beneath collector. In order to
accurately design and improve performance of SCPP, the other parameters like collector slope
and converging-diverging angles of chimney must also be considered. Recently, only initial level
investigations were conducted on aforementioned parameters. Moreover, previously SCPP was
mostly modeled as a 2D (two dimensional) axisymmetric model. Generally, 2D model doesn’t
give us a detailed picture of heat transfer and fluid flow patterns inside the whole system. Hence,
in the present study a three dimensional CFD analysis incorporating solar load, Discrete
Ordinates (DO) and RNG k-ɛ turbulence model at different collector slopes and chimney
diverging angles for Spanish prototype SCPP was performed.
99
Nomenclature
J
Ra
g
L
EP
area (m2)
specific heat (J/kg K)
energy (J)
solar intensity (W/m2)
temperature (K)
AC
C
A
Cp
E
I
T
TE
D
M
AN
U
SC
RI
PT
76
77
78
79
80
81
82
83
84
85
86
87
diffusion flux (kg/ m2s)
Rayleigh number
gravitational acceleration
(m/s2)
length scale (m)
effective viscosity
Greek symbols
α
air velocity (m/s)
β
Poutput power output (W)
t
time (s)
n
refractive index
ε
µ
ρ
v
turbulence kinetic energy due to
buoyancy
source term in momentum equation
heat source in energy equation
effective conductivity
collector’s slope, thermal diffusivity
(m2/s)
chimney diverging angle, thermal
expansion coefficient (K-1)
turbulent dissipation rate (m2/s3)
dynamic viscosity (Pa s)
density (kg/m3)
3
ACCEPTED MANUSCRIPT
̿
λ
overall velocity (m/s)
position vector
direction vector
scattering direction vector
external body force
turbulence kinetic energy
due to mean velocity
gradient
contribution of fluctuating
dilation in compressible
turbulence to overall
dissipation rate
Subscripts
i,j
solid angle (radians)
phase function
scattering coefficient (m-1)
Stefan Boltzmann constant (5.67x
10-8 W/m2 K4)
stress tensor (Pa)
thermal conductivity(W/m-K)
RI
PT
′
`
∅
directional components
SC
pressure (Pa)
kinetic energy (J/Kg)
heat transfer coefficient
(W/m2-k)
axial coordinate (m)
M
AN
U
p
k
ℎ
100
2.
102
2.1 Case study description
103
104
105
106
107
108
109
110
111
112
113
114
115
116
To investigate the effects of varying collector’s slope and chimney diverging angles on
performance of SCPP, Manzanares power plant was considered as a base case and a parametric
analysis was performed by using CFD technique. CFD software ANSYS FLUENT was used for
detailed analysis of heat transfer mechanisms and fluid dynamics involved in SCPP. It was
reported that in Manzanares prototype, to divert the air flow in chimney, height of collector was
raised gradually from 1.85m up to 6m adjacent to the chimney base and turbine was installed
near chimney base on a framework 9m above ground level. Furthermore, a cylindrical shaped
chimney made from thick trapezoid sheets reinforced with rings without any diffuser was
employed in Manzanares plant[1]. The dimensions of this prototype plant along with other
parameters which were used in simulating reference Manzanares case are given in Table1 and
parameters investigated in this study viz. collector slope α and chimney diverging angle β are
presented in Fig.1. After validation of CFD simulation results of Manzanares case, further
numerical simulations were performed at α (collector slope) = 40, 60, 80, 100 and at β (diverging
angle) = 10-30 sequentially to examine their effects on performance of SCPP.
117
2.2
118
119
120
The main governing equations used in modelling SCPP are based on the fundamental mass,
momentum and energy conservation equations described below:
The mass conservation or continuity equation can be written as [17]:
121
Methodology
AC
C
EP
TE
D
101
Mathematical Formulation
+ ∇. = (1)
4
ACCEPTED MANUSCRIPT
Where is the user defined source term or mass added to the continuous phase from dispersed
second phase. In our case the flow is incompressible without any source term, so the above
equation can be simplified as: ∇. = 0 and in Cartesian coordinate system it can be expressed
as:
(
)
*
+
+
=0
(2)
Momentum conservation equation is stated as:
129
130
131
132
133
+
+ ∇. = −∇- + ∇. ̿ + . + (3)
Where - in the above equation is the static pressure, ̿ is stress tensor, . and are the
gravitational body force and external body forces respectively. In rectangular coordinate systems
for an incompressible flow, eq. (3) can be expanded as:
&0
134
/
+ 1'
135
/ + 1'
136
/
&5
&6
+ 1'
&0
'
+ 1)
&0
&0
&5
)
+ 1+
'
+ 1)
&5
)
+ 1+
&5
&6
+ 1)
&6
+ 1+
&6
'
)
+
2 = -/
+
2 = -/
SC
127
128
'
RI
PT
&
126
30,0
'
+
35,0
+
35,5
)
+
36,0
+
2−
+ .'
(4)
)
+
36,5
+
2 − ) + .)
(5)
M
AN
U
122
123
124
125
+
30,5
2 = -/
'
30,6
'
+
35,6
)
+
36,6
+
7
'
7
2−
7
+
+ .+
(6)
In our case, to model buoyancy driven flow, gravitational acceleration was activated in negative
y-direction. Therefore, the terms .' and .+ can be omitted from equations 4 and 6
respectively.
141
2.2.1 Turbulence modeling
142
143
In SCPP air flow is induced due to density variations caused by heat addition. The strength of
such buoyancy driven flows is calculated by Rayleigh number (Ra).
TE
D
137
138
139
140
89 =
151
152
(7)
AB
Flow is fully turbulent if 89 > 10 . Previous studies show that flow in SCPP is turbulent with
89 > 1010 [10]. Consequently, to model turbulence in flow, Renormalization group (RNG) k-ɛ
model was used owing to its better reliability and accuracy to solve a wider class of flows. The
transport equations of (RNG) k-ɛ model are given by [17]:
149
150
:;∆=>? @
10
AC
C
145
146
147
148
EP
144
+
I +
I
'
C
'C
=
D
=
D
'E
'E
FGJ
FG
' H
E
J
' H
E
+ + − I − + J
+ KLJ + KMJ − KNJ JO
− 8J + J
(8)
(9)
For incompressible flows above equations can be simplified as:
+ ' C
D
=
'E
FG
' H
E
+ + − I − + (10)
5
ACCEPTED MANUSCRIPT
158
159
160
161
162
163
164
165
C
D
=
'E
FGJ
J
' H
E
J
+ KLJ + KMJ − KNJ
JO
− 8J + J
(11)
Where 9PQ represent the generation of turbulence kinetic energy due to mean velocity
gradient and buoyancy respectively. 9PQ J are user defined source terms and G , GJ stand for
inverse effective Prandtl numbers for k and ɛ respectively. may be defined as: =
AE
A @
−TTTTTTTT
and for ideal gases = −.D U
, where W is turbulent Prandtl number for
R′ S′
'C
@7VU 'C
RI
PT
157
+ ' I
energy and .D is the component of gravitational vector in Xth direction. The effect of buoyancy
(
on I is determined by KMJ , which can be calculated as: KMJ = Y9Pℎ Z&Z where is the component
of flow velocity parallel to gravitational vector and 1 is the component of flow velocity
perpendicular to gravitational vector. The main difference between RNG and Standard k-ɛ model
lies in the additional term 8J given by: 8J =
^
^_
L` ;] ?
∁\ @] ? L
ɛO
, where c ≡ /ɛ, cf = 4.38, ∁A =
SC
154
155
156
J
b
0.0845 9PQ k = 0.012. Moreover, the model constants KLJ and KNJ in eq. (11) are derived
analytically from RNG theory having values 1.42 and 1.68 respectively.
M
AN
U
153
2.2.2 Modeling Heat transfer
167
168
169
170
171
172
In modeling a solar chimney power plant, we have to consider all three modes of heat transfer
i.e. conduction, convection and radiation. The radiation effects of sun rays entering in
computational domain were calculated by solar load model using Solar Ray Tracing algorithm.
Ray tracing computes the heat flux resulting from incident solar radiations and results are
coupled to ANSYS FLUENT calculations via a source term in energy equation. FLUENT solves
energy equation in the following form [17]:
n + ∇. on p = ∇. /o ∇q − ∑s ℎs ts p + o̿ . p2 + (12)
176
177
178
179
180
181
182
183
184
185
186
187
188
The first three terms on right hand side in the above equation, represent energy transfer due to
conduction, species diffusion and viscous dissipation. indicates volumetric heat sources that
user has defined. is the effective conductivity and ts is the diffusion flux of species j. As
EP
174
175
our flow is incompressible and there are no diffusion terms, eq. (12) can be simplified as:
∇. on p = ∇. ∇q + ̿ . + (13)
Heat transfer through radiations was computed by applying the discrete ordinates (DO) model,
because of the fact that DO model has the ability to solve radiations through semitransparent
walls and a wide range of optical thicknesses [17]. The radiative transfer equation for “DO”
model is:
zx
v= w
v
Q ′
∇. u , + 9 + u , = 9PN
+ y { u /, ′2 ∅, ′
(14)
AC
C
173
TE
D
166
x
zx f
Where I is solar intensity, and represent position vector and direction vector respectively. T
denotes the local temperature and ′ is scattering direction vector. Where ∅ and ′ represent
phase function and solid angle respectively. Further detail about these models can be found in
[17].
6
ACCEPTED MANUSCRIPT
2.3
Computational domain, boundary conditions and solution methodsModel of
190
191
192
193
SCPP was created in ANSYS Design Modeler by using structural dimensions of first working
model of SCPP built in Manzanares Spain given in Table1. To reduce the computational efforts,
the entire domain was divided into half by 1800 revolve. Three dimensional model of SCPP
exhibiting different zones is presented in Fig.2.
194
195
196
197
Computational domain of solar chimney power plant was discretized into hexahedral mesh by
ANSYS ICEM CFD. To achieve most favorable mesh size, a grid independence study was
performed on five different mesh sizes by judging maximum temperature and velocity as
depicted in Table2.
198
199
200
It can be concluded from the Table2 that, temperature and velocity almost remains invariable
after Grid 3. Therefore, Grid 3 was chosen for all simulations as further mesh refinement will
only increase run time.
201
202
The main boundary conditions specified for numerical simulations are depicted in Fig.2 and their
details are given below:
203
204
205
206
207
208
209
210
Sun direction vector was computed from solar calculator by specifying the North, East direction
vectors along Y-axis and X-axis respectively. Similarly, solar irradiation value was calculated
from solar calculator by specifying global position of Manzanares mentioned in Table1. A
typical day of June and day time of 1:00 p.m. was selected for numerical simulations. Although,
atmosphere is not uniform and air properties vary with altitude away from earth surface, but for
smaller altitude as in Manzanares case with chimney height =195m, the air properties can be
assumed to be constant. Thus, in this study values of air properties close to ground were used.
Operating pressure of 1atm and environmental temperature of 303K were selected.
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
Collector inlet is set as pressure-inlet boundary and chimney outlet is set as pressure-outlet
boundary condition. Adopting previous researches, inlet and outlet pressure were set at 0 Pa,
because in modeling natural convection problems, there should be no pressure difference
between inlet and outlet [10,13,19,20]. Total temperature = 303K was defined for pressure inlet
and pressure outlet boundary condition, which is assumed to be equal to static temperature when
flow is incompressible. An average operating density of air = 1.164 kg/m3 was specified in fluid
domain. Flow direction was specified as “normal to boundary” in direction specification method
and absolute reference frame was selected. K and Epsilon method was chosen to compute the
turbulence parameters. Collector roof was considered to be made up of glass having
transmissivity = 0.8, Cp = 840(J/kg-K), ρ = 2700(kg/m3) and λ = 0.78 (W/m-K)[10]. As glass
retains infra-red radiations but allows visible radiations to pass through it, therefore glass
absorptivity and transmissivity for visible radiations was set to 0.04 and 0.92 respectively in
solar boundary conditions. Similarly, glass absorptivity and transmissivity for infra-red
radiations were assigned value of 0.95 and 0.05 respectively. Collector was modeled as a semitransparent wall with mixed boundary condition which accounts for heat transfer by both
convection and solar irradiance. Soil was selected as a ground material having properties as
follows: Cp = 2200(J/kg-K), ρ = 1900(kg/m3) and λ = 1.83 (W/m-K) [10]. Ground was modeled as
AC
C
EP
TE
D
M
AN
U
SC
RI
PT
189
7
ACCEPTED MANUSCRIPT
an opaque wall with mixed boundary condition to model the heat transfer by convection and long
wavelength radiations. Moreover, chimney was considered as an adiabatic wall and symmetry
boundary condition was assigned to center halved zone. These boundary conditions were
adopted for all the cases of sloped collector and diverging chimney.
232
233
234
235
236
237
238
All simulations were performed by assuming steady state condition. Air was treated as an ideal
gas and SIMPLE algorithm was selected as the pressure-velocity coupling scheme. In order to
find accurate results, second order accuracy was selected for all variables. Solution was run up to
2000 iterations to ensure that solution is converged and all residual values reach convergence
criteria. Convergence was also assured by monitoring convergence history of mass flow rate at
outlet shown in Fig. 3, which shows that mass flow rate has stabilized and ensures solution has
converged.
239
3. Results and Discussion
240
3.1
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
To verify the applicability of numerical models adopted in predicting SCPP performance, CFD
simulation results of base case with geometric parameters of Manzanares prototype were
compared with the reported experimental results. Velocity and static temperature contours along
symmetry for base case are presented in Fig.4 and Fig.5 respectively.
Fig.4 depicts that velocity of air increases slowly along the collector’s length but escalates
sharply to 9.16 m/s at the chimney base; owing to the maximum velocity achieved, chimney base
can be considered as a suitable location for turbine installation. It can also be observed from
Fig.5 that, temperature of air rises along the collector’s length due to falling solar radiations and
then almost remains constant along the chimney because, almost negligible heat transfer takes
place through the chimney walls. Temperature of ground is much higher as compared to
collector and reaches up to 318K. This higher temperature of ground is due to the greenhouse
effect and high specific heat of ground. These results demonstrate that SCPP can work at night as
well since, ground acts as a natural heat storage layer capturing heat from solar radiations during
day time and then releasing it later at night.
In Manzanares prototype, air velocity equal to 9m/s and temperature rise of 17-19K were
reported [2]. In CFD simulations at chimney inlet, air velocity = 9.1m/s and temperature rise of
15K beneath collector is predicted. Hence, CFD simulations results are found to be within
acceptable range of experimental results, which ensures the correctness of numerical models
adopted in predicting the SCPP performance.
261
3.2
262
263
264
To analyze the effects of varying collector’s slope on flow distribution of air inside SCPP,
velocity streamlines starting from inlet at different collector slopes are presented in Fig.6 (a-d).
M
AN
U
AC
C
EP
TE
D
Validation of base case
SC
RI
PT
228
229
230
231
Effects of collector’s slope variations on SCPP performance
8
ACCEPTED MANUSCRIPT
It can be observed from the Fig.6 that velocity magnitude increases with an increase in collector
slope reaching 9.05m/s, 9.34m/s, 9.46m/s and 9.75m/s at 40,60,80 and 100 collector slopes
respectively. But simultaneously, it can also be noticed that air flow remains smooth and uniform
until slope = 60 but at higher slopes = 80 and 100, vortices and recirculation of air is detected
beneath collector which obstruct the air flow and may bring down the overall performance. The
reason can be traced down in the fact that as the gap between ground and collector is enhanced;
the gravitational and buoyancy forces become strong and impose diverse manipulation on the air
flow. Fig.7 and Fig.8 evidently reveal the causes of afore mentioned behavior. Fig.7 depicts the
variations of ground surface temperature at different collector’s slopes.
289
290
291
292
So, it can be inferred that at collector’s slope greater than 60, recirculation of air starts developing
beneath the collector due to density gradients formed by uneven distribution of temperature. This
may cause obstruction of air flow within the chimney and deteriorate the performance of wind
turbine.
293
3.3
294
295
296
297
298
299
300
301
302
303
Broadly speaking, SCPP can be viewed as a converging diverging nozzle. Collector inlet to
chimney inlet portion can be regarded as converging portion because; air gets converged into
chimney due to buoyancy effect. Similarly, chimney base can be regarded as throat and upper
portion of chimney can be conceived as diverging section. So, a careful design of SCPP can
deliberately enhance the performance of SCPP. Usually, the converging diverging nozzles are
utilized in applications like rocket engines in which chemical thermal energy generated in
combustion chamber is converted into kinetic energy and supersonic speed has to be achieved.
But in SCPP there is no such source and requirements as those required in rockets. Manzanares
base case can be regarded as a straight chimney having diverging angle (β) = 00. Therefore, to
illuminate the effects of considering SCPP as a nozzle, initial investigations were performed by
RI
PT
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
EP
TE
D
M
AN
U
SC
It can be observed from Fig.7 that ground temperature rises with increasing collector’s slope,
which is due to fact that inclined collector can receive more solar radiations as compared to flat
collector resulting in enhanced heat transfer. Moreover, due to tilted collector some parts of
collector now receives greater amount of solar radiations causing uneven distribution of
temperature. Fig.8 shows the density contours plot against velocity vectors over symmetry plane
for α =100. It indicates that at higher collector’s slope, density gradients are developed beneath
the collector due to uneven distribution of temperature which stimulates the recirculation of air.
The reason behind the increment in velocity with slope is that, in modeling buoyancy driven
flows inside a closed domain, the solution depends upon the mass conserved inside [17]. When
collector’s slope is increased the mass of air confined in it increases as well, resulting in
enhanced mass flow rate as depicted in Fig.9. Another reason is the better mixing of air due to
turbulence effects, resulting in enhanced heat transfer. As in SCPP, main driving force is the
temperature difference of air beneath collector and ambient air, therefore an increment in
velocity is observed with increase in collector’s slope but at the expense of flow uniformity.
AC
C
Effects of diverging chimney on performance of SCPP
9
ACCEPTED MANUSCRIPT
304
305
varying chimney diverging angles (β) = 10-30 while keeping all other parameters same as
Manzanares case.
306
In SCPP, power available from turbine can be expressed as [20]:
Poutput = 0.5 ρAv3
307
(8)
Above expression shows that power is directly proportional to velocity cube and area swept by
turbine. In SCPP, turbine is considered as shrouded turbine whose area depends upon the
chimney area. So, if chimney area is reduced we’ll also have to reduce the turbine area
accordingly, which will bring down the overall power output. So, chimney design requires very
precise calculations for optimum power output.
313
314
315
316
317
318
The distribution of velocity for three different chimney’s diverging angles 10, 20 and 30 is
evaluated in Fig.10 (a-c); the chimney inlet area and all other parameters were kept constant
except the chimney outlet area. The results show that employing a diverging chimney leads to
remarkable increase in velocity. Maximum velocity of 11.6m/s is achieved for case when
diverging angle is 10, while it reaches 11.3m/s and 10.3m/s for diverging angles of 20 and 30
respectively.
319
320
321
322
323
Fig.11 illustrates the static pressure variations along chimney length at different diverging angles
of chimney. It can be observed that static pressure drops dramatically at the chimney base due to
eminent air flow. Furthermore, it can also be noticed that when diverging chimney is used, static
pressure at chimney base drops significantly in accord with Bernoulli’s principle and reaches a
minimum value of -75 Pa when chimney diverging angle =10.
324
325
326
327
328
329
330
331
332
Likewise, Fig. 12 shows variations of dynamic pressure at different diverging angles of chimney.
It indicates that in case of Manzanares prototype plant with straight chimney, a maximum
dynamic pressure of 46 Pa is achieved near chimney base, which drops slowly and reaches 25 Pa
along the chimney. While in the case when a diverging chimney is employed, dynamic pressure
increases significantly at chimney base and then falls abruptly due to expansion of area. A
maximum dynamic pressure of 75 Pa is achieved in case when diverging angle = 10. This
pressure gradient created due to diverging chimney assists in higher air intake and thus a large
amount of electricity can be generated from pressure energy of air by installing a properly
designed pressure staged turbine.
333
334
335
336
337
In order to clearly visualize the effects of chimney diverging angles on air flow inside the whole
system, the velocity streamlines starting from inlet are shown in Fig. 13(a-c). Results show that
at diverging angle β =10, flow remains smooth and uniform but at higher diverging angles of 20
and 30 velocity of air drops drastically after passing through chimney base due to diverging
section and flow separation is detected.
338
339
Finally, the power output for four cases is plotted in Fig.14 and corresponding power increment
is presented in Table 3. As expected, owing to the maximum velocity, power output is highest
AC
C
EP
TE
D
M
AN
U
SC
RI
PT
308
309
310
311
312
10
ACCEPTED MANUSCRIPT
for case β = 10 with remarkable percentage increment of 108%. As power from turbine is directly
related to cube of velocity, a small increment in velocity results in a tremendous increase in
power.
343
344
345
346
Although it was shown that, diverging chimney improves the system performance but, from
constructional point of view, building up a diverging chimney would require special expertise
and may increase the cost of SCPP.
347
4.
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
In this study effects of collector’s slope and diverging chimney on performance of SCPP are
numerically investigated. The scope of this study comprises the validation of CFD simulations
performed on Manzanares prototype with reported experimental results. The validated CFD
approach incorporating RNG k-ɛ, solar load, DO radiation model is then used to investigate the
effects of different collector’s slopes and chimney diverging angles on air flow and heat transfer
in SCPP, aiming to improve its performance. Numerical results show that by increasing
collector’s slope there is a gradual increase in air velocity. But at collector slopes higher than 60,
air flow does not remain uniform and recirculation of air, vortices are detected beneath the
collector due to density gradients formed by uneven distribution of temperature, which may
obstruct the air from entering inside chimney and thus may reduce the overall performance of
SCPP. Likewise, the effects of diverging chimney on performance of SCPP were also
investigated at diverging angles (β) =10-30. Results show that employing a chimney diverging
angle =10 results in a noteworthy increment in velocity form 9.1m/s to 11.6m/s which is
estimated to increase the power output by 108%. Therefore, as compared to other approaches of
increasing the performance of SCPP like increasing height of chimney and collector diameter, by
employing diverging chimney we can achieve high power output at much smaller chimney
height and collector diameter.
365
Acknowledgements
366
367
368
The authors express acknowledgements to the faculty of U. S. Pakistan Centre for Advanced
Studies in Energy (USPCAS-E) NUST Pakistan, who made their valuable contributions and
supported this study.
RI
PT
340
341
342
AC
C
EP
TE
D
M
AN
U
SC
Conclusions
369
370
References
371
372
[1]
W. Haaf, K. Friedrich, G. Mayr, and J. Schlaich, “Solar Chimneys Part I : Principle and
Construction of the Pilot Plant in Manzanares,” no. August 2013, pp. 37–41, 2007.
373
374
[2]
W. Haaf, “Solar Chimneys Part II: Preliminary Test Results from the Manzanares Pilot
Plant,” Int. J. Sol. Energy, vol. 2, pp. 141–161, 2007.
375
[3]
Y. . Dai, H. . Huang, and R. . Wang, “Case study of solar chimney power plants in
11
ACCEPTED MANUSCRIPT
Northwestern regions of China,” Renew. Energy, vol. 28, no. 8, pp. 1295–1304, Jul. 2003.
376
[4]
K. S. Ong and C. C. Chow, “P erformance of a solar chimney,” vol. 74, pp. 1–17, 2003.
378
379
[5]
M. a. dos S. Bernardes, a. Voß, and G. Weinrebe, “Thermal and technical analyses of
solar chimneys,” Sol. Energy, vol. 75, no. 6, pp. 511–524, Dec. 2003.
380
381
[6]
J. P. Pretorius and D. G. Kröger, “Critical evaluation of solar chimney power plant
performance,” Sol. Energy, vol. 80, no. 5, pp. 535–544, May 2006.
382
383
[7]
A. J. Gannon and T. W. Von Backstro, “With System Loss and Solar,” vol. 122, no.
August, pp. 133–137, 2000.
384
385
[8]
M. M. Padki and S. A. Sherif, “On a simple analytical model for solar chimneys,” vol.
349, no. June 1998, pp. 345–349, 1999.
386
387
[9]
R. Sangi, M. Amidpour, and B. Hosseinizadeh, “Modeling and numerical simulation of
solar chimney power plants,” Sol. Energy, vol. 85, no. 5, pp. 829–838, May 2011.
388
389
390
[10]
E. Gholamalizadeh and M.-H. Kim, “Three-dimensional CFD analysis for simulating the
greenhouse effect in solar chimney power plants using a two-band radiation model,”
Renew. Energy, vol. 63, pp. 498–506, Mar. 2014.
391
392
393
[11]
S. K. Patel, D. Prasad, and M. R. Ahmed, “Computational studies on the effect of
geometric parameters on the performance of a solar chimney power plant,” Energy
Convers. Manag., vol. 77, pp. 424–431, Jan. 2014.
394
395
396
[12]
M. Tingzhen, L. Wei, X. Guoling, X. Yanbin, G. Xuhu, and P. Yuan, “Numerical
simulation of the solar chimney power plant systems coupled with turbine,” Renew.
Energy, vol. 33, no. 5, pp. 897–905, May 2008.
397
398
399
[13]
E. S. Nia and M. Ghazikhani, “Numerical investigation on heat transfer characteristics
amelioration of a solar chimney power plant through passive flow control approach,”
Energy Convers. Manag., vol. 105, pp. 588–595, Nov. 2015.
400
401
[14]
P. Guo, J. Li, Y. Wang, and Y. Wang, “Numerical study on the performance of a solar
chimney power plant,” Energy Convers. Manag., vol. 105, pp. 197–205, Nov. 2015.
402
403
[15]
A. R. Shahreza and H. Imani, “Experimental and numerical investigation on an innovative
solar chimney,” Energy Convers. Manag., vol. 95, pp. 446–452, May 2015.
404
405
406
[16]
E. Gholamalizadeh and M.-H. Kim, “CFD (computational fluid dynamics) analysis of a
solar-chimney power plant with inclined collector roof,” Energy, vol. 107, pp. 661–667,
Jul. 2016.
407
[17]
408
[18]
409
410
[19]
411
412
[20]
AC
C
EP
TE
D
M
AN
U
SC
RI
PT
377
“ANSYS FLUENT Theory Guide,” vol. 15317, no. November, pp. 724–746, 2011.
T. D. Canonsburg, ANSYS FLUENT User â€TM s Guide, vol. 15317, no. November. 2011.
P. Guo, J. Li, and Y. Wang, “Numerical simulations of solar chimney power plant with
radiation model,” Renew. Energy, vol. 62, pp. 24–30, Feb. 2014.
A. Rezaei and H. Imani, “Experimental and numerical investigation on an innovative solar
chimney,” Energy Convers. Manag., vol. 95, pp. 446–452, 2015.
413
414
415
12
ACCEPTED MANUSCRIPT
416
417
418
RI
PT
419
420
421
422
SC
423
424
M
AN
U
425
426
427
428
432
433
434
435
436
437
EP
431
AC
C
430
TE
D
429
438
439
440
441
13
ACCEPTED MANUSCRIPT
Figures with captions
M
AN
U
SC
RI
PT
442
443
Fig.1- Schematic diagram of SCPP
444
445
AC
C
EP
TE
D
446
447
448
Fig.2- Computational domain of SCPP with different zones
449
14
ACCEPTED MANUSCRIPT
M
AN
U
SC
RI
PT
450
451
452
Fig. 3- Convergence history of mass flow rate at outlet
454
455
AC
C
EP
TE
D
453
Fig.4- Base case-velocity contours along symmetry
15
M
AN
U
SC
RI
PT
ACCEPTED MANUSCRIPT
456
457
Fig.5- Base case-temperature contours along symmetry
458
AC
C
EP
TE
D
459
460
461
462
Fig.6 (a) - Velocity streamlines at α =40
463
16
ACCEPTED MANUSCRIPT
M
AN
U
SC
RI
PT
464
465
Fig.6 (b) - Velocity streamlines at α =60
467
468
AC
C
EP
TE
D
466
Fig.6 (c) - Velocity streamlines at α =80
17
M
AN
U
SC
RI
PT
ACCEPTED MANUSCRIPT
469
Fig.6 (d) - Velocity streamlines at α =100
470
471
472
AC
C
EP
TE
D
473
474
475
Fig.7- Temperature of ground at varying collector slopes
18
ACCEPTED MANUSCRIPT
M
AN
U
SC
RI
PT
476
477
Fig.8- Effects of density variations on air flow
478
479
290
280
275
270
265
260
AC
C
Mass flow rate(kg/s)
285
EP
295
TE
D
480
255
250
245
240
Manzanares
481
482
slope=4°
slope=6°
slope=8°
slope=10°
Fig.9- Computed mass flow rates at different collector slopes
483
19
M
AN
U
SC
RI
PT
ACCEPTED MANUSCRIPT
484
Fig.10 (a) – Distribution of velocity at β =10
485
AC
C
EP
TE
D
486
487
488
Fig.10 (b) – Distribution of velocity at β= 20
489
20
M
AN
U
SC
RI
PT
ACCEPTED MANUSCRIPT
490
Fig.10 (c) – Distribution of velocity at β= 30
491
493
494
AC
C
EP
TE
D
492
Fig.11- Static pressure at different diverging angles
21
M
AN
U
SC
RI
PT
ACCEPTED MANUSCRIPT
495
Fig.12- Dynamic pressure at different diverging angles
497
498
AC
C
EP
TE
D
496
Fig.13 (a) - Velocity streamlines inside SCPP at β= 10
499
22
M
AN
U
SC
RI
PT
ACCEPTED MANUSCRIPT
500
Fig.13 (b) - Velocity streamlines inside SCPP at β= 20
502
503
AC
C
EP
TE
D
501
Fig.13 (c) - Velocity streamlines inside SCPP at β = 30
504
23
ACCEPTED MANUSCRIPT
505
80
506
70
50
RI
PT
Power (kW)
60
40
30
10
0
Manzanares
508
510
516
517
518
519
520
EP
515
AC
C
514
TE
D
511
513
β=3°
Fig.14- Effect of diverging chimney on power output
509
512
β=2°
M
AN
U
507
β=1°
SC
20
521
522
24
ACCEPTED MANUSCRIPT
Tables
523
524
525
Table 1- Dimensions and parameters used in predicting performance of SCPP
Value
38.990 N
RI
PT
Parameters
Latitude of Manzanares, Spain
3.370 W
Longitude of Manzanares, Spain
GMT of Manzanares, Spain
+2
Height of chimney
195 m
122 m
SC
Radius of collector
Radius of Chimeny
5m
2m
M
AN
U
Collector inlet to ground distance
Distance of Chimney base from Ground
526
527
Table 2- Grid Independence Study
531
Grid
Name
Maximum
velocity(m/s)
1
Grid 1
66619
45.009
8.510
2
Grid 2
114636
45.427
8.638
3
Grid 3
163496
45.28
9.163
4
Grid 4
208024
44.931
9.104
5
346862
44.868
9.108
Cells Count
TE
D
S. No
Grid 5
AC
C
530
Maximum
Temperature(C)
EP
528
529
6m
Table 3- Maximum velocity magnitudes at different chimney diverging angles
Case Description
Maximum
Velocity (m/s)
Percentage increase in Power
(%)
Manzanares (β= 00)
9.1
0
β= 10
11.6
108%
β= 20
11.3
90%
β= 30
10.3
46%
532
25
ACCEPTED MANUSCRIPT
Highlights
EP
TE
D
M
AN
U
SC
RI
PT
CFD analysis of solar chimney power plant is presented.
Effects of sloped collector and diverging chimney are analyzed.
Above optimum collector slope, vortices develop and performance deteriorates.
Employing diverging chimney causes tremendous increase in updraft velocity.
AC
C
•
•
•
•