See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/262374804

CFD Simulation of Deep-Bed Paddy Drying Process and Performance

Article in Drying Technology · May 2014

DOI: 10.1080/07373937.2013.875561

CITATIONS READS
85 2,040

3 authors:

Mohsen Ranjbaran Bagher Emadi

Illumina Saskatchewan Polytechnic
27 PUBLICATIONS 491 CITATIONS 104 PUBLICATIONS 1,434 CITATIONS

SEE PROFILE SEE PROFILE

Dariush Zare
Shiraz University
60 PUBLICATIONS 1,294 CITATIONS

SEE PROFILE

All content following this page was uploaded by Dariush Zare on 06 December 2015.

The user has requested enhancement of the downloaded file.

This article was downloaded by: [New York University]
On: 14 May 2014, At: 15:26
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Drying Technology: An International Journal

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/ldrt20

CFD Simulation of Deep-Bed Paddy Drying Process and

Performance
a a b
Mohsen Ranjbaran , Bagher Emadi & Dariush Zare
a
Biosystems Engineering Department , Ferdowsi University of Mashhad , Mashhad , Iran
b
Biosystems Engineering Department , Shiraz University , Shiraz , Iran
Published online: 14 May 2014.

To cite this article: Mohsen Ranjbaran , Bagher Emadi & Dariush Zare (2014) CFD Simulation of Deep-Bed Paddy Drying
Process and Performance, Drying Technology: An International Journal, 32:8, 919-934, DOI: 10.1080/07373937.2013.875561

To link to this article: http://dx.doi.org/10.1080/07373937.2013.875561

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Drying Technology, 32: 919–934, 2014
Copyright # 2014 Taylor & Francis Group, LLC
ISSN: 0737-3937 print=1532-2300 online
DOI: 10.1080/07373937.2013.875561

CFD Simulation of Deep-Bed Paddy Drying Process and

Performance
Mohsen Ranjbaran,1 Bagher Emadi,1 and Dariush Zare2
1
Biosystems Engineering Department, Ferdowsi University of Mashhad, Mashhad, Iran
2
Biosystems Engineering Department, Shiraz University, Shiraz, Iran

sources in the world. The quality of paddy during storage

Computational Fluid Dynamics (CFD) was applied three- mainly depends on its moisture content.[3] Drying technol-
dimensionally to simulate the drying behavior of paddy in a ogies are used to decrease the high moisture content of
deep-bed dryer. The commercial CFD software Fluent 6.3.26 was paddy to enhance its shelf-life during storage. The
used. The deep-bed paddy drying process and performance were
Downloaded by [New York University] at 15:26 14 May 2014

deep-bed drying method is commonly used to dry deep

studied by incorporating user-defined function (UDF) in Fluent writ-
ten in C language. The predicted drying parameters were compared layers of grains like paddy.
with experimental data of deep-bed drying of paddy. The values of Computer simulation is a cheap, less time-consuming
mean relative deviation (MRD), standard error of prediction (SEP), and energy-saving method to obtain detailed and near-
and maximum error of prediction (MEP) for prediction of grain precise information about food process operations. Today,
moisture content, air temperature, and absolute humidity were less access to high-speed computers has enhanced the ability to
than 6, 10, and 9%; 0.33% (d.b), 1.24
C, and 0.06% (kg/kg of
dry air); and 2.25% (d.b), 6.8
C, and 0.37% (kg/kg of dry air), apply computational fluid dynamics (CFD) for simulations
respectively, which reflect reasonable accuracy. Moreover, the ener- of complex engineering problems. Jamaledine and Ray[4]
getic and exergetic performance of deep-bed paddy drying were have reviewed the CFD simulations of several drying pro-
simulated and analyzed. The effects of inlet air temperature and cesses and systems. Although several commercial CFD
mass flow rate on the performance parameters were investigated. packages like Fluent are available, they have still not been
It was shown that the application of higher levels of inlet air tem-
perature and lower mass flow rates yielded higher exergy efficiencies fit for simulations of several food and post-harvest agricul-
of deep-bed paddy drying. tural processes like drying,[5] since they need additional
codes to consider the biological nature of food and agricul-
Keywords CFD simulation; Deep-bed paddy drying; Energy; tural materials. Thus, further work is required to provide
Exergy such codes to expand the ability of commercial CFD soft-
ware in the area of food and biosystems engineering.
Several studies have been done on the simulations of
INTRODUCTION deep-bed drying of grains, among them the works by Zare
Drying is an important unit operation which is widely et al.,[6] Naghavi et al.,[7] and Srivastava and John[8] are of
used to preserve agri-food products. In general, it refers interest. Most of these works used one-dimensional models
to the removal of moisture from a material. It can prolong to simulate the deep-bed drying process. Three-
the shelf-life of products and reduce the costs of transpor- dimensional CFD modeling provides more realistic simula-
tation and storage. Drying is one of the most energy- tions of deep-bed dryers=silos. It can be applied to simulate
intensive operations with high importance in several and analyze deep-bed dryers=silos with multidirectional
industries. Dincer[1] believes that the drying industry can aeration and non-isotropic porous zones.
be considered as the heart for industrial systems. Thus, Dincer[9] defined exergy as the maximum amount of
several investigations have been carried out on the drying work which can be produced by a stream of matter, heat,
processes of food materials with the aims of economical or work as it comes to equilibrium with a reference
considerations, environmental concerns, and product qual- environment. The applications of the exergy concept in
ity aspects.[2] Rice is one of the most important food the design and control of drying systems can result in better
efficiency, better cost-effectiveness, better design and analy-
Correspondence: Mohsen Ranjbaran, Biosystems Engineering sis, better environment, and better sustainability.[1] Accord-
Department, Ferdowsi University of Mashhad, P.O. Box
9177948974, Mashhad, Iran; E-mail: mo.ranjbaran@stu.
ing to Dincer and Sahin,[10] the exergy analysis can be
um.ac.ir applied to design more efficient thermal systems by reduc-
Color versions of one or more of the figures in the article can ing their inefficiencies. Aghbashlo et al.[11] recently
be found online at www.tandfonline.com/ldrt.

919
 920 RANJBARAN ET AL.

provided a review that summarizes the using of exergy  To analyze the energetic and exergetic perfor-
analysis in drying operations. They also identified mance of deep-bed paddy drying
prospects for future research in this area. Icier et al.[12] per-  To validate the CFD drying model using experi-
formed exergetic analysis on the performance of broccoli mental results
drying in three different drying systems. They investigated  To develop user-defined functions (UDFs) and
the effect of drying air temperature on the exergy efficiency, macros to be used in the Fluent (Fluent Inc.,
exergy destruction, and the improvement potential of dry- Lebanon, NH) software
ing processes. Aghbashlo et al.[13] carried out energy and
exergy analyses of drying of potato slices in a semi-
MATERIALS AND METHODS
industrial continuous-band dryer. They showed that the
feeding rate and temperature and flow rate of drying air In the present model, it was assumed that grains were
had important effects on the energy and exergy consump- in temperature equilibrium with inter-granular drying air
tion during drying. Prommas et al.[14] carried out energy (i.e., grains and inter-granular air were at the same
and exergy analyses in the hot-air drying process of porous temperature[19]), which can be seen in the work of
media. They studied the effects of particle size on the Thorpe.[20] Moreover, similar to the experimental deep-bed
performance parameters. paddy drying apparatus of Zare et al.,[6] which was used to
Although several investigations have been performed on validate the results of the present CFD simulations,
the energy and exergy analysis of drying processes based on the walls in the computational geometry were considered
Downloaded by [New York University] at 15:26 14 May 2014

experimental data, little attention has been devoted to the adiabatic.

simulation of energetic and exergetic performance of dry-
Governing Equations
ing processes. Such simulations can effectively contribute
to better design, management, control, and greenization Momentum Transfer
of drying systems. They can help designers analyze the The continuity equation for drying air is[19]
thermodynamic performance of drying systems prior to

scale-up. Ranjbaran and Zare[15] worked on the simulation @qa
þ r: qa*u ¼ 0 ð1Þ
of energetic and exergetic performance of microwave- @t
assisted fluidized bed drying of soybeans. Their results were
obtained using a previously validated mathematical model where qa is air density and *u is air velocity vector.
reported by Zare and Ranjbaran.[16] They showed that The momentum conservation equation for the drying air
computer simulations can effectively be used in the perfor- is written as follows[19]:
mance analysis of a drying process. Assari et al.[17]

performed energy and exergy analysis for batch fluidized @ qa*u
 ¼
bed drying of wheat based on the Eulerian two-fluid þ r: qa*u*u ¼ rP þ r: s þqa*g þ Sm ð2Þ
model. They showed that predicted values of energy and @t
exergy efficiencies had good agreement with reported ¼
available correlations. A search of the available litera- where s is the Reynolds stress tensor. The boundary con-
ture revealed that there are few studies which have been ditions for Eq. (2) included no-slip condition at the walls,
done on the exergetic analysis of the performance of prescribed air mass flow rate at the inlet, and prescribed
deep-drying of grains. Amantéa et al.[18] worked on the air pressure at the outlet.
numerical simulation techniques for deep-bed corn drying
efficiency analysis. They applied a numerical meshless Heat Transfer
scheme based on cubic radial basis functions to solve a The heat transfer inside the porous zone was modeled by
nonlinear coupled set of one-dimensional partial differen- the following energy balance equation[20]:
tial equations.
@
* 
CFD modeling techniques can be used in energy and
ðqa e Ia þ qs ð1  eÞIs Þ þ r: qa u Ia ¼ keff r2 T þ Sh ð3Þ
exergy analysis of deep-bed grain drying processes. Such @t
analyses are helpful for engineers by providing a better
insight into deep-bed grain drying. Moreover, they can where e is the bed porosity. The boundary conditions for
yield novel practical ideas to improve, optimize, and more Eq. (3) included adiabatic bed walls and prescribed inlet
effectively control such drying systems. air temperature. keff is the effective thermal conductivity
The objectives of this work were as follows: of bulk grain (W=m.K), which was calculated as:

 To simulate the deep-bed paddy drying process

and performance based on 3D CFD modeling keff ¼ eka þ ð1  eÞks ð4Þ
 CFD SIMULATION OF DEEP-BED PADDY DRYING 921

The enthalpy of air and solids were defined as: where a is the permeability, C2 is the internal resistance
  factor, and v is air velocity in vertical direction. To calcu-
Ia ¼ Cpa þ wCpv T ð5Þ late these two parameters, the relationship reported by
Hunter[21] was used:
Is ¼ Cps T ð6Þ
dP
¼ Rv  Sv2 ð10Þ
where Cpa, Cpv, and Cps are specific heat capacity of air, dy
vapor, and moist solids, respectively. T is the absolute
temperature of drying air and solids. where R and S are constants which are dependent on the
grain type. Thus, the values of a and C2 were calculated
Exergy Balance as below:
As reported by Assari et al.,[17] the exergy balance
equation was derived by multiplying the entropy by T0 (ref- 1 R
¼ ð11Þ
erence ambient temperature) and subtracting the obtained a l
expression from the energy equation.
2S
   C2 ¼ ð12Þ
@   T qa
q e 1a  T0 Cpa þ wCpv ln
@t a T0
Downloaded by [New York University] at 15:26 14 May 2014

  For rough rice (paddy) the values of R and S are

T
þ qs ð1  eÞ Is  T0 Cps ln 1952 Pa  s=m2 and 10419 Pa  s2=m3, respectively. According
T0 to ASABE,[22] the pressure drop in the horizontal direction
   ð7Þ
* T may be regarded as 70% of that in the vertical direction.
þ r: qau Ia  T0 ðCpa þ wCpv Þln
T0 This resulted in a non-isotropic porous zone. Application
   
T0 T 0
of non-isotropic porous zone is more important during
¼ 1 keff r2 T þ 1  Sh  e_ xdest simulation of dryers=silos with multidirectional aeration.[20]
T T
Moisture source term. The moisture source term in the
where e_ xdest is the rate of exergy destruction per unit gas phase due to the moisture evaporation of grains is as[5]:
volume of each cell (W=m3). The second term on the
righthand side of Eq. (7) is the exergy consumption in
@M
evaporation of moisture from grains. Sw ¼ ð1  eÞqs ð13Þ
@t
Moisture Transfer
where e is the porosity of bed, qs is the density of paddy,
To simulate the distribution of moisture inside the and M is grain moisture content (d.b). To predict the dry-
porous bed, moisture was considered as a User-Defined ing rate of paddy in each computational cell, the following
Scalar (UDS). Thus, an additional equation was solved thin-layer drying equation was applied[23]:
to predict the moisture transport in air[5]:
  
t n1 
@M 1
@ ðqa wÞ
   ¼ ðM  Me Þ kn ð14Þ
þ r: qa*
uw ¼ r: qa Deff rw þ Sw ð8Þ @t 3600 3600
@t
with
where w is the absolute humidity of drying air (kg=kg,
d.b.), Deff is the effective moisture diffusion coefficient k ¼ 0:02958  0:4456RH þ 0:01215ðT  273:15Þ
through bulk grain, Sw is the moisture source term in the
gas phase that is due to the moisture evaporation from n ¼ 0:13365 þ 1:93653RH  1:77431RH2
grains. The boundary conditions for this equation included þ 0:009468ðT  273:15Þ
zero moisture flux at walls and prescribed air absolute
humidity at the inlet. where T and RH are the temperature and relative humidity
of drying air, respectively.
Definitions of Source Terms
It should be noted that the predicted values of efficien-
Momentum source term. The resistance of the porous cies are highly dependent on the values of predicted drying
medium to air flow in vertical direction was calculated by[19]: rate. Thus, application of a suitable thin-layer drying equa-
  tion is very important to avoid probable unexpected theor-
dP l 1 etical results such as prediction of efficiency values above
Sm ¼ ¼  v þ C2 qa v2 ð9Þ
dy a 2 100%.
 922 RANJBARAN ET AL.

The following empirical equation reported by Zuritz and l ¼ 1:691  105 þ 4:984  108 ðT  273:15Þ  3:187
Singh[24] can be applied to predict the equilibrium moisture
content (Me) of paddy:  1011 ðT  273:15Þ2 þ1:319  1014 ðT  273:15Þ3
ð24Þ
 c
A   
Me ¼ 0:001 ð15Þ 6887 T
B Pvs ¼ 0:1 exp 27:0214   5:31 ln ð25Þ
T 273:16
A ¼  lnð1  RHÞT
101:3w
 23:438 RH ¼ ð26Þ
7 T 0:62189Psv þ wPvs
B ¼ 2:667  10 1
641:7

1 Performance Analysis
C¼ 5 2:1166 As reported by Ranjbaran and Zare,[15] the energy
4  10 T
efficiency of the drying process is defined as the ratio of
the energy demanded in evaporation of moisture from
Energy source term. The energy source term due to the
grains to the energy incorporated in the inlet air. The
effect of evaporative cooling can be considered as[5,20]:
energy consumed in evaporation was calculated in all com-
Downloaded by [New York University] at 15:26 14 May 2014

putational cells of the solution domain. Thus, the energy

@M
Sh ¼ ð1  eÞqs hfg ð16Þ efficiency was defined as:
@t
PN
where hfg is the latent heat of vaporization of water i¼1 ðSh dV Þi
ge ¼ ð27Þ
(J=kg)[5]: _E a; in

hfg ¼ 2503000  2386 ðT  273:15Þ; 273:15 < T ðK Þ < 533:15

Here, dV is the volume (m3) of each computational cell
ð17Þ and N is the number of total cells in the solution domain.
The rate of energy incorporated in inlet air is as:
Property Equations for Paddy   
The moisture-dependent property equations of paddy E_ a;in ¼ m
_ a Cpa þ wCpv Ta;in  T0 ð28Þ
were calculated by the following relationships reported by
Meeso et al.[25]: The exergy efficiency of drying process was defined as:

 PN   
M i¼1 1  TT0 Sh dV i
CPs ¼ 1110 þ 44:8 ð18Þ gex ¼ ð29Þ
M þ1 E_ xa;in
 
M
ks ¼ 0:0863 þ 0:00134 ð19Þ
M þ1

e ¼ 0:623  0:25M ð20Þ

qb ¼ qs ð1  eÞ ¼ 552 þ 282M ð21Þ

Property Equations for Air

The thermodynamic properties of drying air were esti-
mated by the following relationships[5]:

101:325
qa ¼ ð22Þ
0:287T

Cpa ¼ 1009:26  4:0403  103 ðT  273:15Þ þ 6:1759

 104 ðT  273:15Þ2 4:097  107 ðT  273:15Þ3
FIG. 1. Computational geometry, including a plenum chamber and
ð23Þ porous bed.
 CFD SIMULATION OF DEEP-BED PADDY DRYING 923

where E_ xa;in is exergy rate of inlet air: the flow field. The corresponding results are not shown
  for the sake of brevity. Grid-independent results were
  T obtained using a two-block grid with resolutions of
_E xa;in ¼ m
_ a Cpa þ wCpv T  T0  T0 ln : ð30Þ
T0 80  80  50 for bed and plenum chamber. The quality of
the grid was also examined by taking into account the cell
aspect ratio, cell equi-angle skew, and cell growth factor.
Simulations Procedure As reported by Ranjbaran and Zare,[5] the optimum values
The GAMBIT grid generator (Fluent Inc., Lebanon, for these parameters might be regarded to be less than 5,
NH) was used to divide the solution domain (i.e., a 40 cm  0.5, and 1.10, respectively. In the present study, the
40 cm  25 cm bed and a plenum chamber with a 10 cm  maximum value for cell aspect ratio of the structured hexa-
10 cm inlet and height of 22 cm) into structured hexahedral hedral grid was 3, for equi-angle skew it was 0.38, and for
computational cells (see Fig. 1). To ensure obtaining cell growth factor it was 1. By taking into consideration the
grid-independent results, a grid study was performed and Courant number requirements,[5] a time step of 0.01 s was
several grid resolutions were used in the simulations of chosen. The simulations were performed using a

TABLE 1
Description of simulation settings
Downloaded by [New York University] at 15:26 14 May 2014

Description Method=value
Geometry and grid
Geometrical space 3D
Mesh resolution at porous bed 80  80  50
Mesh resolution at plenum chamber 80  80  50
Maximum grid aspect ratio 3
Maximum cell growth factor 1
Maximum equi-angle skew 0.38
Porous bed section area (m2) 0.16
Porous bed height (m) 0.25
Plenum chamber inlet section area (m2) 0.01
Plenum chamber height (m) 0.22
Materials
Paddy bulk density (kg=m3) 608.4
Paddy specific heat capacity (J=kg.K) 1117.5
Paddy thermal conductivity (W=m.K) 0.0866
UDS (moisture) diffusivity in gas phase (kg=m.s) 4.90  107
Operating, boundary, and initial conditions
Operating pressure (Pa) 101325
Gravitational acceleration (m=s2) 9.81
Inlet boundary condition Velocity-inlet
Outlet boundary condition Pressure-outlet, zero gage pressure
Bed wall boundary condition Wall, adiabatic
Viscous resistance of porous zone (1=m2) 10.626  107 (vertical direction), 7.44  107 (horizontal direction)
Inertial resistance of porous zone (1=m) 18605 (vertical direction), 13023 (horizontal direction)
Fluid porosity 0.57
Solution
Solver Pressure based
Discretization method Second-order upwind
Time step (s) 0.01
Unsteady formulation Second-order implicit
Maximum number of iterations per time step 30
Convergence criteria 0.001
Under-relaxation factors 0.5  0.8
Pressure-velocity coupling SIMPLE[29]
 924 RANJBARAN ET AL.

3ddp-varsion of Fluent 6.3.26. Moisture in the gas phase  To calculate the moisture and heat source terms
was considered to be a UDS. Thus, a UDS transport equa-  To calculate the performance parameters
tion was activated in Fluent to calculate the distribution of
Fluent has provided several macros that can be emplo-
absolute humidity of drying air in the bed. A UDF written
yed in writing of UDFs. Table 2 introduces some of these
in C language was prepared to expand the ability of Fluent
macros and their applications. Furthermore, the complete
for this study. The air flow in the porous zone was con-
UDF has also been provided in the Appendix. This code
sidered to be laminar. The standard k-E turbulence
can be modified and used to simulate the deep-bed drying
model[19] was considered at the plenum chamber. The gov-
process and performance of other cereals and granular
erning equations were solved by applying a pressure-based
materials.
segregated solver. The typical values of under-relaxation
factors were 0.5–0.8. The complete simulation settings are
listed in Table 1. Stepwise Procedure to Run the Model
Generally, a 3D CFD simulation of the deep-bed drying
User-Defined Function (UDF) process using a 3ddp-version of Fluent was performed as
The UDF was used for the following aims in each time follows:
step:
 Read the 3D mesh into Fluent (File ! Read !
 To update grain moisture content and calculate Case), check it (Grid ! Check), and scale it
Downloaded by [New York University] at 15:26 14 May 2014

moisture-dependent material properties (Grid ! Scale).

TABLE 2
Selected macros[19] that are useful for writing UDFs to simulate a deep-bed drying process and performance
Macro Application
DEFINE macros
DEFINE_INIT(initial_humidity, d) Introducing initial values of grains moisture content and gas absolute
humidity to Fluent
DEFINE_ADJUST(moisture_update, d) Updating grain moisture content at each time step for each
computational cell according to the temperature and absolute humidity
of inter-granular air; performing energy and exergy analysis at each
time step
DEFINE_SOURCE(moisture_source, d) Calculating the moisture source term (Eq. (13)) and returning it to the
solver
DEFINE_SOURCE(evaporative_cooling, d) Calculating the energy source term (Eq. (16)) and returning it to the
solver
Flow variable macros
C_T (c, t) Returning cell temperature (K)
User-defined scalar and memory macros
C_UDSI (c, t, i) Accessing to the value of ith user-defined scalar (UDS)
C_UDMI (c, t, i) Accessing to the value of ith user-defined memory
Geometry macros
C_VOLUME (c, t) Returning the value of cell volume (m3)
Looping macros
thread_loop_c (t, d) Looping over all cell threads in a domain
begin . . . end_c_loop (c, t) Looping over all cells in a cell thread
Time macros
RP_Get_Real (‘‘flow-time’’) Returning current simulation time (s)
RP_Get_Real (‘‘physical-time-step’’) Returning current time step size (s)
RP_Get_Integer (‘‘time-step’’) Returning the integer number of run time steps
Note that some of the commonly used Fluent data types are as below:.
cell_t, which is a data type for a cell identifier (e.g., cell_t c).
face_t, which is a data type for a face identifier (e.g., face_t f).
Thread, which is a data structure in Fluent (e.g., Thread  t).
Domain, which is the highest-level data structure in Fluent (e.g., Domain  d).
 CFD SIMULATION OF DEEP-BED PADDY DRYING 925

TABLE 3
Experimental drying conditions[26]
Grains initial Inlet air Inlet air Inlet air Temperature of
moisture content, temperature mass flow absolute humidity, surroundings Bed
Test no. d.b (kg=kg) (
C) rate (kg=m2  s) (kg=kg of dry air) (
C) depth (cm)
1 0.2587 45 0.22 0.01 27 25
2 0.2587 50 0.22 0.01 27 25
3 0.2557 50 0.10 0.01 27 25

 Activate a UDS transport equation (Define ! Error of Prediction (SEP), and also the Maximum Error
User-Defined ! Scalars). of Predictions (MEP) for several drying variables.
 Set the number of required User-defined Memor-
ies (Define ! User-Defined ! Memory). 0 !2 112
 Read the concatenated UDF (Define ! 1X n
/exp;j  /pre;j
MRD ¼ @ A 100 ð31Þ
User-Defined ! Functions ! Interpreted) and n j¼1 /exp;j
select each part in its suitable place (i.e., DEFI-
Downloaded by [New York University] at 15:26 14 May 2014

NE_INIT and DEFINE_ADJUST: Define !
P  2 12

n
User-Defined ! Function Hooks, DEFINE_ j¼1 /exp;j  /pre;j
SOURCE: Define ! Boundary Conditions). SEP ¼ ð32Þ
n1
 Set the solver and unsteady formulation (Define
 
! Models ! Solver), activate the energy equation MEP ¼ max /exp;j  /pre;j  ð33Þ
(Define ! Models ! Energy), and set the viscous 1jn

model (Define ! Models ! Viscous).

 Set material properties (Define ! Materials) and where / is an arbitrary variable and n is number of
boundary conditions for inlet, outlet, walls, and measurements for each experiment.
porous zone (Define ! Boundary Conditions).
 Set solution controls including discretization RESULTS AND DISCUSSION
methods, under relaxation factors, and pressure- Model Validation
velocity coupling algorithm (Solve ! Controls The results of verification tests for predictions of grain
! Solution). moisture content, air temperature, and absolute humidity
 Initialize the solution (Solve ! Initialize ! are shown in Table 4. The values of MRD and SEP for pre-
Initialize) and patch initial values to the solution diction of paddy moisture content were less than 6% and
domain (Solve ! Initialize ! Patch). 0.33% (d.b), respectively, compared with experimental data
 Set time step size, number of time steps, and provided by Zare.[26] The MEP for prediction of paddy
maximum iteration per time step and start itera- moisture content was less than 2.25% (d.b). The variations
tions (Solve ! Iterate). of grain moisture content for drying air temperature of
50
C and mass flow rate of 0.22 kg=m2  s, at two different
bed depths of 9 and 18 cm, are illustrated in Fig. 2. As can
Model Validation Procedure be seen in Table 4, the values of MRD, SEP, and MEP for
The simulation results were validated by using some predictions of drying air temperature and absolute
experimental data provided by Zare[26] for deep-bed drying humidity were less than 10 and 9%, 1.24
C and 0.06%
of paddy. The drying conditions are listed in Table 3. These (kg=kg of dry air), and 6.8
C and 0.37% (kg=kg of dry
experimental data include the measured moisture contents air), respectively. Figure 3 shows the variations of pre-
of paddy (in two different bed heights of 9 and 18 cm), the dicted and experimental values of drying air temperature
measured drying air temperature, and absolute humidity during drying time. Moreover, the variations of air absol-
(in three different bed heights of 9, 18, and 25 cm) for dif- ute humidity are illustrated in Fig. 4. The absolute
ferent drying conditions. The detailed information regard- humidity of inlet air for all drying conditions was
ing experimental procedures, sample preparation methods, 0.01 kg=kg (see Table 3). After 30 min, the values of air
deep-bed dryer apparatus, and measuring techniques and absolute humidity were about 0.0136 and 0.0172 kg=kg at
instruments can be obtained in Zare and Chen.[27] The vali- bed depths of 9 and 25 cm, respectively. Due to the
dations of CFD simulations were carried out by calculation reduction of paddy moisture content, the absolute
of the Mean Relative Deviation (MRD), the Standard humidity of air decreased during drying. Similar trends
 926 RANJBARAN ET AL.

TABLE 4
Results of verification tests
Paddy moisture Air absolute humidity
content (%, d.b) Air temperature (
C) (%, kg=kg of dry air)
Depth Depth Depth Depth Depth Depth Depth Depth
Drying condition 9 cm 18 cm 9 cm 18 cm 25 cm 9 cm 18 cm 25 cm
MRD(%)
45
C and 0.22 kg=m2  s 4.39 4.21 8.03 8.67 8.30 7.45 7.18 7.55
50
C and 0.22 kg=m2  s 3.81 5.72 7.39 8.69 8.42 8.08 8.49 7.32
50
C and 0.10 kg=m2  s 5.03 5.21 9.36 8.45 6.16 7.27 8.09 7.00
SEP
45
C and 0.22 kg=m2  s 0.232 0.243 1.145 1.158 1.133 0.036 0.039 0.035
50
C and 0.22 kg=m2  s 0.220 0.323 1.020 1.168 1.142 0.038 0.057 0.030
50
C and 0.10 kg=m2  s 0.249 0.281 1.238 1.155 0.605 0.030 0.045 0.022
MEP
45
C and 0.22 kg=m2  s 1.55 1.34 6.55 6.65 2.64 0.14 0.37 0.29
Downloaded by [New York University] at 15:26 14 May 2014

50
C and 0.22 kg=m2  s 1.35 2.24 5.02 5.33 5.11 0.13 0.23 0.19
50
C and 0.10 kg=m2  s 1.41 1.90 5.07 6.80 3.28 0.14 0.27 0.20

were also obtained for other drying conditions. The results errors in measurement of experimental values of grain
showed that the model was capable of predicting grain moisture content, air temperature and absolute humidity,[6]
moisture content, drying air temperature, and absolute and probable errors in estimation of pressure drop of air
humidity with a reasonable accuracy.[28] The deviations passing through the bed.
of predicted drying variables from corresponding experi-
mental data can be due to several sources, such as making
Simulation of Drying Process
the assumption that grains were in temperature equilibrium
with inter-granular air (this may yield errors in prediction Variations in paddy moisture content towards the bed
of grain temperature, especially in the early stages of dry- center line for different drying conditions are provided
ing; however, after some time they can be assumed to be in Fig. 5. As expected, a higher drying air temperature
at the same temperature[19]), probable errors in prediction resulted in lower final grain moisture content after
of drying rates using the thin-layer drying model (for 150 min of drying time. This is due to the fact that by appli-
example, the thin-layer drying equation may not be so cation of higher temperatures the vapor pressure gradient
accurate for that particular variety of paddy), and also between inside grains and outside surrounding increases,
physical and thermal properties of materials, probable which favors higher moisture removals. Moreover, applying

FIG. 2. Variation in paddy moisture content predicted by CFD model FIG. 3. Variation in drying air temperature predicted by CFD model vs.
vs. experimental data[26] at inlet air temperature of 50
C and mass flow experimental data[26] at inlet air temperature of 50
C and mass flow rate of
rate of 0.22 kg=m2  s. 0.22 kg=m2  s.
 CFD SIMULATION OF DEEP-BED PADDY DRYING 927

During deep-bed drying of grains, a non-uniform drying

takes place. The grains at the top region of the bed may
absorb moisture transported by humid air, especially dur-
ing the first drying period. Thus, their moisture content
may increase to values higher than their initial moisture
content. This phenomenon takes longer when applying
lower mass flow rates and deeper beds.
A typical contour of paddy moisture content after
55 min drying with inlet air temperature of 50
C and mass
flow rate of 0.22 kg=m2  s is provided in Fig. 6. The contour
explains the non-uniform, three-dimensional distribution
of paddy moisture content during deep-bed drying. Such
three-dimensional distributions were also predicted for
FIG. 4. Variation in drying air absolute humidity predicted by CFD other drying variables. Due to the rapid change in the
model vs. experimental data[26] at inlet air temperature of 50
C and mass cross-section of the plenum chamber, the air approaching
flow rate of 0.22 kg=m2  s.
the bed had higher velocity at the center. This led to higher
vapor pressure gradient between inside grains and the out-
higher mass flow rate of inlet air resulted in higher drying side surroundings. Thus, the grains at the center of the bed
Downloaded by [New York University] at 15:26 14 May 2014

rates. Similar results have also been reported by other were dried with higher drying rates.
researchers.[6,8] Figure 7 illustrates the variations in air temperature dis-
tribution towards the bed center line for different drying
conditions. At drying air temperature of 50
C, applying a
lower level of mass flow rate (i.e., 0.10 kg=m2  s) resulted
in a cooler bed. As predicted by the model, the temperature
distribution in the bed is closely linked with the value of
grain moisture content. In the early stages of drying, the
effect of evaporative cooling was higher due to higher dry-
ing rates, especially for drying condition of 50
C and
0.10 kg=m2  s, such that at t ¼ 5 min the temperature of
grains located on the top region of the bed decreased to
about 25
C. Meanwhile, the variation in the rate of energy
consumption in evaporation of moisture from grains is
shown in Fig. 8. In the early stages of drying, energy was

FIG. 5. Variations in paddy moisture content towards the bed center line FIG. 6. Contours of paddy moisture content after 55 min of drying with
for different drying conditions. inlet air temperature of 50
C and mass flow rate of 0.22 kg=m2  s.
 928 RANJBARAN ET AL.

The absolute humidity of inlet air was 0.01 kg=kg. It

increased as it passed through the moist grains bulk. This
can be seen in Fig. 9. Applying higher levels of air tempera-
ture increased the absolute humidity of drying air due to
higher evaporation of moisture from grains. Moreover,
lower mass flow rates of inlet air increased the inter-
granular air absolute humidity. The rate of moisture
transfer from grains into air, towards the bed depth, is illu-
strated in Fig. 10. In the first period of the drying process,
the rate of moisture transfer from grains into air has nega-
tive values in the top region of the bed. It means that these
grains were absorbing moisture from humid air. This
phenomenon is more severe when applying lower levels
of air mass flow rate and deeper beds. The predicted results
showed that in the early stages of drying the evaporation
mainly took place at the bottom region of the bed. Then,
after some time, the rate of moisture transfer from grains
into air at upper regions increased due to an increase in
Downloaded by [New York University] at 15:26 14 May 2014

the available enthalpy required for evaporation.

The pressure drops of air passing through the bed calcu-
lated by Eq. (9) were about 178 Pa and 61 Pa when mass

FIG. 7. Variations in air temperature towards the bed center line for
different drying conditions.

mainly consumed in evaporation from grains that were

located at the bottom of the bed. As time goes on, by
reduction in the moisture contents of bottom grains, the
available enthalpy at higher regions increased and more
drying took place there.

FIG. 8. Variation in the rate of energy consumption for evaporation of

moisture from paddy at inlet air temperature of 50
C and mass flow rate FIG. 9. Variations in air absolute humidity towards the bed center line
of 0.10 kg=m2  s towards the bed center line. for different drying conditions.
 CFD SIMULATION OF DEEP-BED PADDY DRYING 929

of drying (for example, the relative humidity exceeded

0.92 Pa=Pa for drying condition of 50
C and 0.10 kg=
m2  s after 5 min of drying). Application of higher mass
flow rates of inlet air resulted in lower relative humidity.
A higher mass flow rate favored the moisture removal from
inter-granular pores and decreased the inter-granular
vapor pressure and, therefore, the relative humidity. The
results showed that the relative humidity of inter-granular
air was more closely linked with the mass flow rate of inlet
air than its temperature.

Performance Analysis
FIG. 10. Variation in the rate of moisture transfer from paddy into air at In addition to the prediction of the deep-bed drying pro-
inlet air temperature of 50
C and mass flow rate of 0.10 kg=m2  s towards cess of paddy, the energetic and exergetic performance of dif-
the bed center line. ferent drying conditions were also simulated. Figure 12
illustrates the variations in energy and exergy efficiencies
flow rates of drying air were 0.22 and 0.10 kg=m2  s, for two different levels of inlet air temperature and at mass
Downloaded by [New York University] at 15:26 14 May 2014

respectively. Figure 11 shows the variations in relative flow rate of 0.22 kg=m2  s. At the initial stage of the drying
humidity of inter-granular air towards the bed center line process, the air temperature declined to less than reference
for different drying conditions. The relative humidity of ambient temperature (27
C). Thus, the value of exergy
inter-granular air was higher at the top regions of the efficiency at this stage was calculated as a negative value.
bed, especially for lower mass flow rates in the early stage

FIG. 12. Variation in energy and exergy efficiencies of deep-bed drying

FIG. 11. Variation in relative humidity of inter-granular air towards the of paddy at mass flow rate of 0.22 kg=m2  s and two different levels of inlet
bed center line for different drying conditions. air temperature.
 930 RANJBARAN ET AL.

A rapid increase in the energy and exergy efficiencies was

due to high drying rates at the early stage of the drying per-
iod. This means that a large portion of inlet enthalpy was
consumed in moisture evaporation from grains, which
caused efficiencies to reach an instantaneous local peak.
Such phenomenon has also reported by Amantéa et al.[18]
After this stage, due to a decrease in drying rate, the energy
and exergy efficiencies took a downward trend. As reported
by Ranjbaran and Zare,[15] the applications of first and
second laws of thermodynamics for determination of the
performance of drying processes might lead in different
results. Within the framework of the first law of thermody-
namics, a higher temperature difference between inlet air
and reference ambient yields lower energy efficiencies.
Thus, as can be seen in Fig. 12, a higher level of inlet air
temperature resulted in lower energy efficiency. On the
other hand, within the framework of the second law of
thermodynamics, in addition to the temperature difference
Downloaded by [New York University] at 15:26 14 May 2014

between inlet air and the reference ambient, the level of

temperature of inlet air and reference ambient are also con-
sidered in calculations of efficiency. The second law of ther-
modynamics recommended the application of higher levels
of inlet air temperature to reach higher exergetic efficien-
cies. Such results have also been reported by Amantéa
et al.[18] using one-dimensional simulation of the perfor-
mance of deep-bed drying of corn.
The variations in energy and exergy efficiencies of
deep-bed drying of paddy at two different levels of inlet
air mass flow rates are illustrated in Fig. 13. The results FIG. 13. Variation in energy and exergy efficiencies of deep-bed drying
showed that both the first and second laws of thermo- of paddy at inlet air temperature of 50
C and two different levels of inlet
dynamics recommended the applications of lower levels air mass flow rate.
of inlet air mass flow rate to enhance energy and exergy
efficiencies, respectively. When a higher level of inlet air capabilities to predict mold production during drying and
mass flow rate was applied, grains had less time to consume aeration, contributing to better design and management
this amount of energy to evaporate moisture. Thus, a larger of industrial deep-bed grain dryers and silos.
portion of inlet energy went out of the bed. This yielded
lower energy=exergy efficiencies. However, application of CONCLUSIONS
low inlet air mass flow rates resulted in an increase in In this article, the capability of commercial CFD soft-
paddy moisture content and relative humidity of drying ware Fluent was expanded to be used in simulation of
air at higher regions of the bed (see Fig. 11). In such the deep-bed drying process and performance of cereal
regions, grains are at the risk of quality degradation due grains. The drying process of paddy in a deep-bed dryer
to higher values of water activities in the bed, which may was simulated three-dimensionally. The simulation results
lead to mold production. This is more important for were compared with some experimental data of deep-bed
industrial-scale, deep-bed paddy drying processes which drying of paddy. The obtained values of mean relative devi-
need longer drying times. Thus, to choose a proper level ation (MRD), standard error of prediction (SEP), and
of inlet air mass flow rate, both thermodynamic efficiency maximum error of prediction (MEP) for prediction of
and grain quality should be taken into account. grain moisture content, air temperature, and absolute
This work tried to expand the ability of commercial humidity were less than 6, 10, and 9%; 0.33% (d.b),
CFD software Fluent to be used in simulation of the 1.24
C, and 0.06% (kg=kg of dry air); and 2.25% (d.b),
deep-bed paddy drying process and performance. In addi- 6.8
C, and 0.37% (kg=kg of dry air), respectively. The
tion to the energy concerns, the quality of the product simulation results showed that to reach higher exergetic
should also be considered. Mold production is an impor- efficiencies, a higher inlet air temperature and lower mass
tant issue for industrial deep-bed grain drying and storage. flow rate should be applied. However, the risk of grain
CFD techniques have the potential to enhance our quality degradation should be taken into account during
 CFD SIMULATION OF DEEP-BED PADDY DRYING 931

the choice of inlet air temperature and mass flow rate. e Equilibrium
Further work is needed to equip commercial CFD software exp Experimental
for prediction of probable mold production inside grains in Inlet
during several drying and aeration conditions. pre Predicted
s Solids
NOMENCLATURE v Vapor
C2 Inertial resistance factor, 1=m
Cp Specific heat capacity, J=kg K
Deff Effective moisture diffusion coefficient
ACKNOWLEDGMENTS
through bulk grain, m2=s
The authors would like to acknowledge Ferdowsi
d.b Dry basis
University of Mashhad (FUM) for all its support of this
E_ Energy rate, W
work.
E_ x Exergy rate, W
e_ xdest Rate of exergy destruction per unit volume
of each cell, W=m3
REFERENCES
g Gravity acceleration, m=s2
1. Dincer, I. Exergy as a potential tool for sustainable drying systems.
I Fluid enthalpy, J=kg Sustainable Cities and Society 2011, 1, 91–96.
hfg Latent heat of vaporization of water,
Downloaded by [New York University] at 15:26 14 May 2014

2. Okos, M.R.; Narsimhan, G.; Singh, R.K.; Weitnauer, A.C. Food

J=kg dehydration. In Handbook of Food Engineering; Heldman, D.R.,
ka Thermal conductivity of air, W=m K Lund, D.B., Eds.; Marcel Dekker: New York, 1992.
ks Thermal conductivity of paddy, W=m K 3. Hacıhafızoglu, O.; Cihan, A.; Kahveci, K. Mathematical modelling of
drying of thin layer rough rice. Food and Bioproducts Processing 2008,
keff Effective thermal conductivity, W=m K 8(6), 268–275.
M Grain moisture content, d.b., kg=kg 4. Jamaleddine, T.J.; Ray, M.B. Application of computational fluid
m _ Mass flow rate of air, kg=m2 s dynamics for simulation of drying processes: A review. Drying
Me Equilibrium moisture content of paddy, Technology 2010, 28, 120–154.
d.b., kg=kg 5. Ranjbaran, M.; Zare, D. CFD modeling of microwave-assisted
fluidized bed drying of moist particles using two-fluid model. Drying
MRD Mean relative deviation Technology 2012, 30(4), 362–376.
MEP Maximum error of prediction 6. Zare, D.; Minai, S.; Mohamad Zadeh, M.; Khoshtaghaza, M.H.
P Pressure, Pa Computer simulation of rough rice drying in a batch dryer. Energy
Pvs Saturated vapor pressure, kPa Conversion and Management 2006, 47, 3241–3254.
RH Relative humidity of gas, Pa=Pa 7. Naghavi, Z.; Moheb, A.; Ziaei-Rad, S. Numerical simulation of rough
rice drying in a deep-bed dryer using non-equilibrium model. Energy
Sh Energy source term, W=m3 Conversion and Management 2010, 51, 258–264.
Sm Momentum source term, Pa=m 8. Srivastava, V.K.; John, J. Deep bed grain drying modeling. Energy
Sw Gas moisture source term, kg=m3 s Conversion and Management 2002, 43, 1689–1708.
SEP Standard error of prediction 9. Dincer, I. Thermodynamic, exergy and environmental impact. Energy
T Temperature, K (
C if specified) Sources 2000, 22, 723–732.
10. Dincer, I.; Sahin, A.Z. A new model of thermodynamic analysis of a drying
t Time, s process. International Journal of Heat and Mass Transfer 2004, 47, 645–652.
*
u Velocity vector, m=s 11. Aghbashlo, M.; Mobli, H.; Rafiee, S.; Madadlou, A. A review on
v Air velocity in vertical direction, m=s exergy analysis of drying processes and systems. Renewable and
w Absolute humidity of gas, d.b., kg=kg Sustainable Energy Reviews 2013, 22, 1–22.
Greek Letter 12. Icier, F.; Colak, N.; Erbay, Z.; Kuzgunkaya, E.H.; Hepbasli, A. A
comparative study on exergetic performance assessment for drying
a Permeability, m2 of broccoli florets in three different drying systems. Drying Technology
q Density, kg=m3 2010, 28, 193–204.
e Gas porosity, m3=m3 13. Aghbashlo, M.; Kianmehr, M.H.; Arabhosseini, A. Energy and exergy
¼
s Reynolds stress tensor, Pa analyses of thin-layer drying of potato slices in a semi-industrial con-
l Gas viscosity, Pa s tinuous band dryer. Drying Technology 2008, 26, 1501–1508.
14. Prommas, R.; Rattanadecho, P.H.; Cholaseuk, D. Energy and
dV Cell volume, m3 exergy analyses in drying process of porous media using hot air.
ge Energy efficiency International Communications in Heat and Mass Transfer 2010, 37,
gex Exergy efficiency 372–378.
/ Arbitrary variable 15. Ranjbaran, M.; Zare, D. Simulation of energetic and exergetic perfor-
mance of microwave-assisted fluidized bed drying of soybeans. Energy
Subscripts 2013, 59, 484–493.
16. Zare, D.; Ranjbaran, M. Simulation and validation of microwave-
0 Reference surrounding or ambient assisted fluidized bed drying of soybean. Drying Technology 2012,
a Air 30(3), 236–247.
 932 RANJBARAN ET AL.

17. Assari, M.R.; Basirat Tabrizi, H.; Najafpour, E. Energy and exergy 24. Zuritz, C.A.; Singh, P. An equation to compute the heat of evapor-
analysis of fluidized bed dryer based on two-fluid modeling. Inter- ation of water for rough rice during drying. Drying Technology
national Journal of Thermal Sciences 2013, 64, 213–219. 1985, 3(3), 421–435.
18. Amantéa, R.P.; Fortes, M.; Martins, J.H.; Ferreira, W.R. Numerical 25. Meeso, N.; Nathakaranakule, A.; Madhiyanon, T.; Soponronnarit, S.
simulation techniques for optimizing thermodynamic efficiencies of Modelling of far-infrared irradiation in paddy drying process. Journal
cereal grain dryers. Drying Technology 2013, 31, 672–683. of Food Engineering 2007, 78, 1248–1258.
19. Fluent 6.3, User Manual; Fluent, Inc.: Lebanon, NH, 2006. 26. Zare, D. Computer simulation of rough rice drying in a deep bed
20. Thorpe, G.R. The application of computational fluid dynamics codes batch dryer. Ph.D. thesis, Tarbiat Modares University, Tehran, Iran,
to simulate heat and moisture transfer in stored grains. Journal of 2006.
Stored Products Research 2008, 44, 21–31. 27. Zare, D.; Chen, G. Evaluation of a simulation model in predicting
21. Hunter, A.J. Pressure difference across an aerated seed bulk for some the drying parameters for deep-bed paddy drying. International
common duct and store cross-sections. Journal of Agricultural Journal of Computers and Electronics in Agriculture 2009, 68,
Engineering Research 1983, 28, 437–450. 78–87.
22. American Society of Agricultural and Biological Engineers. Resistance 28. Brooker, D.B.; Bakker-Arkema, F.W.; Hall, C.W. Drying and
to Airflow of Grains, Seeds, other Agricultural Products, and Perforated Storage of Grains and Oilseeds; Van Nostrand Reinhold: New York,
Metal Sheets, D272.3; ASABE: St. Joseph, MI, 2007. 1992.
23. American Society of Agricultural and Biological Engineers. Thin-Layer 29. Patankar, S.V. Numerical Heat Transfer and Fluid Flow; Hemisphere:
Drying of Agricultural Crops, S448.1; ASABE: St. Joseph, MI, 2006. Washington, DC, 1980.
Downloaded by [New York University] at 15:26 14 May 2014

APPENDIX

= =
= UDF for simulation of Deep-bed Paddy Drying Process and Performance  =
= By: Mohsen Ranjbaran =
= Ferdowsi University of Mashhad (FUM), Mashhad, Iran =
# include ‘‘udf.h"
= definitions of constants & initial values:  =
= M: initial moisture content of paddy, d.b., (kg=kg).  =
= RHO: paddy true density (kg=m3).  =
= E: bed void fraction (decimal).  =
= Tref: reference ambient temperature (K).  =
= Tin: initial temperature of paddy (K)  =
= Ein: energy rate of inlet air (W)  =
= Ex_in: exergy rate of inlet air (W)  =
# define M —
# define RHO —
# define E —
# define Tref —
# define Tin —
# define E_in —
# define Ex_in —
int last_timestep ¼ 1;
DEFINE_INIT(initial_humidity, d)
f
real Tabs, w, a, b, C, Pvs, RH;
Thread  t;
cell_t c;
thread_loop_c (t, d)
f
begin_c_loop(c, t)
f
C_T(c, t) ¼ Tin;
C_UDMI(c, t, 0) ¼ M;
Tabs ¼ C_T(c, t);
 CFD SIMULATION OF DEEP-BED PADDY DRYING 933

Pvs ¼ 0.1  exp(27.0214 - (6887 = Tabs) - (5.31  log (Tabs = 273.16))); = Eq (25) =
b ¼ 2.667e-7  pow((1-(Tabs=641.7)),-23.438);
C ¼ 1 = (4e5  pow(Tabs,-2.1166));
a ¼ b  pow (1000  M, 1=C);
RH ¼ 1 - exp (- a = Tabs); = Eq (15) =
if (RH > 0.99)
f RH ¼ 0.99;g
w ¼ (Pvs  RH  0.62189)=(101.3 - Pvs  RH); = Eq (26) =
C_UDSI(c, t, 0) ¼ w;
g
end_c_loop(c, t)
g
g
DEFINE_ADJUST (moisture_update, d)
f
real n_ts,m,time,dMdt, Tabs,RH,mr,t_hour,Psat,time_step,k,power, a,b,C,Me,Mnew, w, Pvs,E_evp;
real Ex_evp, V,n, Hfg, etta_E, etta_Ex,t_E_evp, t_Ex_evp,d_E_evp,d_Ex_evp;
Thread  t;
Downloaded by [New York University] at 15:26 14 May 2014

cell_t c;
n_ts ¼ RP_Get_Integer (‘‘time-step’’);
time_step ¼ RP_Get_Real (‘‘physical-time-step’’);
time ¼ RP_Get_Real (‘‘flow-time’’);
t_hour ¼ time=3600;
if (last_timestep ! ¼ n_ts)
f
last_timestep ¼ n_ts;
thread_loop_c (t, d)
f
begin_c_loop(c, t)
f
m ¼ C_UDMI(c, t, 0);
w ¼ C_UDSI(c, t, 0);
Tabs ¼ C_T(c, t);
Pvs ¼ 0.1  exp(27.0214 - (6887 = Tabs) - (5.31  log (Tabs = 273.16))); = Eq (25) =
RH ¼ (101.3  w)= (Pvs (0.62189 þ w)); = Eq (26) =
if (RH > 0.99)
f RH ¼ 0.99;g
C_UDMI(c, t, 1) ¼ RH;
a ¼ -1  Tabs  log (1-RH);
b ¼ 2.667e-7  pow((1-(Tabs=641.7)),-23.438);
C ¼ 1 = (4e5  pow(Tabs,-2.1166));
Me ¼ 0.001  pow((a=b),C); = Eq (15) =
k ¼ 0.02958-(0.4456 RH)þ(0.01215 (Tabs - 273.16));
n ¼ 0.13365 þ (1.93653 RH)-(1.77431 pow(RH, 2))þ( 0.009468 (Tabs - 273.16));
dMdt ¼ (m - Me)  (- k  n  pow(t_hour, (n-1)))  pow(3600, -1); = Eq (14) =
Mnew ¼ (time_step  dMdt) þm;
C_UDMI(c, t, 0) ¼ Mnew;
C_UDMI(c, t, 2) ¼ dMdt;
V ¼ C_VOLUME (c, t);
Hfg ¼ - 2386  (Tabs - 273.16) þ 2503000; = Eq (17) =
E_evp ¼ - RHO  (1-E)  V  dMdt  Hfg;
if (E_evp <0)
f E_evp ¼ 0;g;
Ex_evp ¼ E_evp (1-(Tref=Tabs));
C_UDMI(c, t, 3) ¼ E_evp;
 934 RANJBARAN ET AL.

C_UDMI(c, t, 4) ¼ Ex_evp;
t_E_evp þ ¼ E_evp;
t_Ex_evp þ ¼ Ex_evp;
g
end_c_loop(c, t)
d_E_evp þ ¼ t_E_evp;
d_Ex_evp þ ¼ t_Ex_evp;
g
etta_E ¼ 100  d_E_evp = E_in; = Eq (27) =
etta_Ex ¼ 100  d_Ex_evp = Ex_in; = Eq (29) =
printf(‘‘Energy efficiency ¼ %g nn’’,etta_E);
printf(‘‘Exergy efficiency ¼ %g nn’’,etta_Ex);
g
g
DEFINE_SOURCE (moisture_source, c, t, dS, eqn)
f
real Sw, dMdt;
Downloaded by [New York University] at 15:26 14 May 2014

dMdt ¼ C_UDMI(c, t, 2);

Sw ¼ - dMdt  (1-E)  RHO; = Eq (13) =
dS[eqn] ¼ 0.0;
C_UDMI(c, t, 5) ¼ Sw;
return Sw;
g
DEFINE_SOURCE (evaporative_cooling, c, t, dS, eqn)
f
real Sh,dMdt, Hfg, Tabs;
Tabs ¼ C_T(c, t);
dMdt ¼ C_UDMI(c, t, 2);
Hfg ¼ - 2386  (Tabs - 273.16) þ 2503000; = Eq (17) =
Sh ¼ dMdt  (1-E)  RHO  Hfg; = Eq (16) =
if (Sh > 0)
f Sh ¼ 0;g
dS[eqn] ¼ 0.0;
C_UDMI(c, t, 6) ¼ Sh;
return Sh;
g

View publication stats