Journal of Food Engineering 144 (2015) 148–155

Contents lists available at ScienceDirect

Journal of Food Engineering

journal homepage: www.elsevier .com/locate / j foodeng

Spatial homogeneity of drying in a batch type food dryer with diagonalair flow design

http://dx.doi.org/10.1016/j.jfoodeng.2014.08.0030260-8774/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Address: Department of Agricultural Engineering, KasselUniversity, Nordbahnhofstr. 1a, 37213 Witzenhausen, Germany. Tel.: +49 5542 981663; fax: +49 5542 98 1520.

E-mail address: waseem_amjad@daad-alumni.de (W. Amjad).

Waseem Amjad a,⇑, Anjum Munir b, Albert Esper c, Oliver Hensel a

a Department of Agricultural Engineering, Kassel University, Witzenhausen, Germanyb Department of Farm Machinery & Power, University of Agriculture, Faisalabad, Pakistanc INNOTECH mbh, Stuttgart, Germany

a r t i c l e i n f o

Article history:Received 23 May 2014Received in revised form 29 July 2014Accepted 3 August 2014Available online 14 August 2014

Keywords:Batch dryerAir distributionDrying uniformityANSYS-Fluent

a b s t r a c t

Batch type dryers are some of the most widespread equipment used for food dehydration. One majordrawback of dryer is the spatial heterogeneity of air distribution in the drying chamber. A new batch typedryer with diagonally airflow inlet channel (along the length of drying chamber) has been developed. Thisfeature caused drying uniformity by exposing the entire food product to incoming warm air. A flow sim-ulation software, ANSYS-Fluent has been used to predict the profile of air distribution in drying chamber.A spatial homogeneity of air distribution was found. To validate the designed geometry of dryer and itsperformance evaluation, experiments were conducted using potatoes (slices, 4 mm thickness) as dryingmaterial. The result, expressed as drying curves for all food buckets, showed high R2 value. The simulatedresults of airflow were compared with experimental measured data. This comparison revealed a goodcorrelation coefficient of 87.09% for airflow distribution.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Batch type dryers are widely used due to simple design require-ment. But problem of uneven air distribution encounter in mostbatch type drying systems. This leads to problems of low dryingefficiency and lack of homogeneity of the products being dried(Mirade, 2003; Misha et al., 2013). Most of the research in batchdrying processes has targeted the effect of temperature and airvelocity as primary influencing parameters for the quality of driedproducts while the lack of uniformity in air flow is a crucial factor.Taking air distribution as primary design parameter has been themain objective in past few research (Adams and Thompson,1985; Ayensu and Asiedu-Bondzie, 1986; Mathioulakis et al.,1998; Kiranoudis et al., 1999; Shawik et al., 2001; Mirade, 2003;Babalis et al., 2005; Margaris and Ghiaus, 2006; Gül ah andCengiz, 2009; Amanlou and Zomorodian, 2010; Jacek et al., 2010;Tzempelikos et al., 2012; Darabi et al., 2013). Therefore, the designof airflow is one of the most important factors in batch type dryers.The process of drying, the drying medium and the geometry of thedrying chamber determine the uniformity of drying and thus thequality of the finished products (Tzempelikos et al., 2012). It means

that after selecting a drying process and drying medium, configu-ration of drying chamber is the critical design parameter. It decidesthe uniform air flow over the product. Non uniformity in air dryingoccurs mostly in drying chamber in fixed bed dryers. Therefore, theuse of baffle plates/air straighter is a common practice to overcomethis problem in large drying chamber (Janjai et al., 2006; Romanet al., 2012). Apart of their benefits to use, they cause not onlyincrease in dryer construction cost but also result in velocity pres-sure drop. So more energy is consumed in drying process.

The present study was conducted to design a new batch typedryer to get uniform drying with uniformly distributed air flowin drying chamber. Normally, in fixed-bed batch dryer, air flowsfrom one side of drying chamber to the other side. Due to this,air gets gradually moist while moving through the product lyingat start, and its drying potential is therefore reduced at the timeof its flow over the product lying at the end (Kröll, 1978; Müllerand Heindl, 2006). So, the inlet side of food buckets within the dry-ing chamber was designed diagonally to maintain velocity pressureas well as to avoid the use of baffle plates. The main concept was tomake possibility for whole drying material to be exposed with theincoming warm air without baffle plates.

The main objectives of this study were:

� To design an air flow pattern with the application of ANSYS Flu-ent in batch type food dryer without baffles in such a way thatthe entire product should be exposed to warm air.

Fig. 2. Diagonal design of the airflow channel in drying chamber. It is line diagramof the drying chamber which shows the diagonal placement of food buckets tomake airflow diagonal at their inlet side.

W. Amjad et al. / Journal of Food Engineering 144 (2015) 148–155 149

� To develop dryer with the designed airflow pattern, evaluatethe design using sample product (potatoes) and comparingthe experimental data of air flow with the simulated data.

2. Description of the dryer

A new batch dryer was designed and developed(11 m � 1.20 m � 1.25 m) with diagonal air flow channel in thedrying chamber at the inlet side of food buckets. It comprised onthree major parts named as connector, lower half and upper halfwhich is also called as drying chamber. Lower half of the dryerwas of the shape of a channel converging towards the lower sideof connector. It included a constant speed axial tube fan (Dia.0.7 m, 453 m3/h, 2.2 kW) for airflow and an electric water–air heatexchanger to heat up the incoming air. A water pump was used forthe circulation of water into the system along with a pressuregauge and thermal expansion valve (to absorb excess water pres-sure). A connector (made of galvanized iron) was used to make aconnection of lower half to the upper half of the dryer (Fig. 1). Inthe upper half of the dryer, twenty-five food buckets (each wasof dimension 0.6 m � 0.4 m � 0.29 m) were arranged diagonallyon a rolling track (for easy loading and unloading of food buckets).Each bucket covered a distance of 0.4 m in the drying chamber.

The inlet sides of these diagonally arranged buckets gave ashape of diagonal airflow channel at an angle of 1.42� with the wallof drying chamber in longitudinal direction (Fig. 2). It is the distinctdesigning feature of the dryer which would cause the flow throughall the buckets evenly. The walls of dryer were made of rock-woolsandwiched into galvanized iron sheets for easy machining andexcellent insulation. Two opening doors (each was of dimension0.65 m � 0.36 m) were kept for the loading and unloading of buck-ets at both sides of the upper half of the dryer. At the outlet side ofbuckets, a rectangular passage was kept just before the outlet door(0.30 m � 0.15 m) for the recirculation of air. This passage was cov-ered and opened with the flap of outlet door. The working of outletdoor (opening/closing time of flap) was controlled with a controllerbased on set temperature. A control panel was used to set temper-ature and time (three different temperatures can be set for threedifferent intervals of drying time).

3. Simulation of the design

3.1. Simulation details

Fluent has been used for simulation purpose in ANSYS work-bench. It provides a comprehensive suite of computational fluid

Fig. 1. 3D model of new designed batch type dryer. The complete geometry of the dr

dynamics (CFD) software for modeling fluid flow and other relatedphysical phenomena (Fluent user’s guide, 2005). The geometry ofthe dryer was modeled and analyzed for two configurations ofair flow channels in the drying chamber namely, straight air flowchannel and diagonal air flow channel (h = 1.42�). The simulationwith straight flow channel was done to assess the suitability ofdiagonal air flow design and to show a comparative change in air-flow regime in the dryer compartment.

To study the profile of air distribution was the concern in sim-ulation, so only the influencing parts of the dryer (connector anddrying chamber) were designed and simulated in 3D model. Thepurpose of 3D model was also to assess flow variation along thedepth of drying chamber as well which was not assessable in 2Dmodel. Keeping in view the importance of air flow pattern, the casewas simulated as steady state condition. For boundary conditions,lower part of the connector was taken as air inlet which receivedheated air coming out from heat exchanger. A value of 5 m/s nor-mal to air inlet was assigned as inlet velocity. k–e standard turbu-lence model was used for air flow turbulence. The characteristicsand settings of simulations are tabulated in Table 1.

The resistance to air flow due to the food layer (single) wasassumed negligible due to its smaller thickness (4 mm thick slices).That is why food was not modeled as solid objects or as porousmedia. Practically, these food layers were at successive distancesin a bucket, so application of porous media was not applicable asfood resistance. The main concern was that if a food bucket getsuniform distribution of air flow, it would facilitate a uniform dry-ing of the material lying in that bucket.

3.2. Simulated results

Fig. 3 shows comparative path lines of velocity for straight (a)and diagonal (b) air flow channels. In straight airflow channel, as

yer is made using Pro. Engineering CAD software. It shows all the parts of dryer.

Table 13D model properties and settings of the simulations.

Number of elements 9,613,901 (straight airflow)9,542,733 (diagonal airflow)

Volume main body 4.36667 m3

Grid type 3D, tetrahedral, unstructuredTurbulence model k–e standardDiscretization Second-order upwindWall friction model No slipInlet air velocity (normal to air inlet) 5.0 m s�1

Mass flow inlet 1.31688 kg s�1

Outlet pressure 0 Pa

150 W. Amjad et al. / Journal of Food Engineering 144 (2015) 148–155

the air entered in the upper half of the dryer (drying chamber) itmoved directly to the end of drying chamber. It caused higher airvelocity at the end of drying chamber especially through the lastbucket (avg. 7 m/s) comparative to other parts of drying chamberwhere zones of higher and lower airflow had been established. Italso caused the variation in static pressure at the walls of foodbuckets.

On the other hand, the design with diagonal airflow channelshowed a good distribution of incoming air through the buckets(Fig. 3b). In this arrangement, the air converged as it movedtowards the end of drying chamber due to its diagonal flow. Itresulted almost same velocity pressure at the inlet side of buckets(inlet flow channel) and it caused air to flow through all the foodbuckets evenly. At the outlet side of buckets (outflow channel),velocity path lines showed the occurrence of almost an equal out-flow from each bucket. It strengthens the design for uniform airdistribution along the length of drying chamber.

In order to visualize air distribution along the depth of dryingchamber (along the height of buckets) as well, planes were drawnat three positions (bottom, middle and top) of the drying chamber.It is more effective way to get through the analysis of airdistribution.

Fig. 4 shows velocity contours at three different positions ofdrying chamber for both straight (Fig. 4a) and diagonal (Fig. 4b)airflow designs. Air velocity decreased as it entered into the buck-ets, so the velocity scale was decreased to 6 ms�1 to assess the vari-ations clearly in airflow through the buckets.

In straight airflow design, velocity contours at all three posi-tions of drying chamber showed a higher flow at the end of dryer,

Fig. 3. Path lines of air velocity for straight (a) and diagonal (b) airflow cases. It showsstraight airflow (Fig. 3a) and diagonal airflow (Fig. 3b) designs.

same as it was observed with velocity path lines in a whole dryingchamber.

Velocity contours for diagonal airflow design at three positionsshowed a uniform distribution, same as it was observed withactual velocity scale (Fig. 3). As air entered into a bucket, turbu-lence created at bucket’s wall. This turbulence occurred especiallyfrom the bottom side of the bucket’s wall, some percentage ofentered air hit the front wall of bucket, creating a swirl and thenstated to move in the bucket’s region. It can be observed fromthe velocity contours of the planes at three positions for diagonalairflow design which showed uniformity of air flow not onlythrough the length of dryer but along the heights of buckets as well(Fig. 4b).

So simulated analysis for diagonal airflow pattern gave gooduniformity of air distribution through the concerned region (foodbuckets).

4. Performance evaluation of the dryer

4.1. Drying trials

To proceed drying trials, the dryer was instrumented to assessits working (Fig. 5). Air velocity was measured with TA5 hotwireanemometer with a resolution of 0.01 m s�1 at two positions oflower half of the dryer (after heat exchanger & at the start of con-nector) and at four positions of upper half of the dryer (at start,middle and at the end of diagonal air flow channel and at theoutlet).

Temperatures at every one meter distance of drying chamberwere measured by inserting thermocouples (K-type ± 1.5 k) at thetop of drying chamber, linked with data logger (Agilent 34970A,Malaysia) to assess temperature distribution throughout the dry-ing chamber. Mini data logger (MSR-145 ± 2%, Swiss) was usedboth at inlet and outlet positions of drying chamber to measurerelative humidity. Electric weight balance (BIZERBA, Germany)was used for weighting of fresh and dry material for drying ratemeasurements at different intervals of an experiment.

The experiments were carried out using locally available pota-toes. Preliminary trials were conducted with single and double lay-ers of potatoes in each bucket to asses drying trend along thelength of drying chamber. After that number of layers wasincreased to four in each bucket to assess the drying uniformity

comparison of airflow distribution using ANSYS Fluent between non-diagonal i.e.

Fig. 4. Contours of air velocity at bottom, middle and top positions of drying chamber for straight (a) and diagonal (b) airflow cases. It shows the air flow distribution alongthe height of drying chamber.

W. Amjad et al. / Journal of Food Engineering 144 (2015) 148–155 151

not only along the length of dryer but also along the height of eachbucket (among four trays). These layers were built up by puttingperforated trays (0.1 cm in thickness) within each bucket at suc-cessive intervals (0.058 m).

Four perforated trays in a bucket were loaded with fourkilogram potatoes slices (4 mm); each of them was carrying onekilogram of drying material. These potatoes were sliced with anadjustable cutting thickness cuter (Bosch MAS62). The operating

Fig. 5. Fabricated new dryer and testing apparatus. It shows the physical setup ofthe developed dryer.

152 W. Amjad et al. / Journal of Food Engineering 144 (2015) 148–155

drying temperature (48 �C) was set with controller. Potatoes weredried up to twelve percent final moisture contents. During experi-ment, periodic weighing method was used for each bucket tocalculate its drying rate.

In case of four layers of drying material, the dryer was not fullyloaded due to difficulty in handling of material during experimentsand to avoid material cost for repetitive trials. So, the drying trialswere divided into five sets of experiments each comprised of fivebuckets. Each set (five buckets) covered a distance of 2 m in thedrying chamber, so five trials covered the entire length of dryingchamber. It was tried to make sure the same drying conditionsfor all the trials at their start. Therefore, the drying material wasnot loaded for each trial just after the start of dryer until therequired air conditions were obtained within the drying chamber.

4.2. Drying results

Fig. 6 shows the experimental results in the form of dryingcurves plotted for each set of trial separately to show spatialuniformity in drying. Although, these were found quite smoothat various distances in drying chamber but during constant rateof drying, there were slight variations in moisture contents amongbuckets. It can be observed in first (Fig. 6a), fourth (Fig. 6d) andfifth (Fig. 6e) set of trials comparative to second (Fig. 6b) and third(Fig. 6c) experimental trials. As air entered into the drying cham-ber, it moved towards the end of dryer and built a velocity pressurewithin the airflow channel due to its diagonal design. Air flow wasnot so smooth just after its entrance into the airflow channel due topushing pressure exerted by incoming air. So, the velocity pressuresettled more smoother at the middle of airflow channel. That iswhy there were found slight span of moisture contents in firstset of trials (Fig. 6a) after second and fourth hours of its drying timecomparative to other. Similarly, at the end of airflow channel, dueto comparative less air velocity, small variation in moisture con-tents were found among buckets during constant rate of drying(Fig. 6d and e). The observed small variations during constant rateof drying occurred because of surface moisture which is looselybounded and its evaporation rate significantly varies with slightvariation in air flow. But all the drying material in each set ofexperimental trial came to almost at an equal final moisture con-tents during the phase of second falling rate. This was the mainobjective of the study that whole the product, lying throughoutthe length of drying chamber, should come at same drying levelwith minimum variation at the end of drying time. The dryingmaterial in all the buckets was being exposed to incoming warmair at same time, so it facilitated the occurrence of almost uniformdrying. For each set of drying data, sigmoid logistic model wasfound best for curve fitting with high R2 value using

sigmaplot-12 as shown in Fig. 6. It showed that a smooth dryingtrend had been shown by all the food buckets under same dryingconditions. Many factors affect the quality of dried product likemedium of drying, tray material, pre treatment processes etc butdrying temperature is more critical. A part of product dries fasterwill exhibit different quality than that which dries slower. Unifor-mity of drying is achieved with uniform distribution of heat.Because zones with different airflow cause heterogeneity in dryingwhich ultimately reduces dryer efficiency by increasing energyconsumption and drying time (Kiranoudis et al., 1999; Romanet al., 2012). As convective heat transfer (hc) of a product varieswith its position in the drying chamber and drying time, so varia-tion in drying rates effect quality parameters of drying materials.No doubt drying temperature is important to maintain the qualityof dried product as some products are heat sensitive. But importantis to distribute that temperature uniformly, so that entire productat all places of drying chamber show same changes in their studiedquality parameters. In this design of diagonal airflow, the productwas dried with uniform drying rate, so it would result uniformquality drying for other concerning quality parameters at all placesof drying chamber. Because uniform quality drying is linked withuniform air distribution. This design of airflow provided quiteencouraging results by eliminating the factor of zones of overand under drying.

Fig. 7 shows the comparative drying trend of all the buckets atvarious intervals of drying time. Overall it shows the drying behav-ior of food material along the entire length of the drying chamber.The variation of moisture contents among the food buckets at var-ious intervals of drying was observed quite low especially as dryingproceeded towards its falling rate.

Experimental measurements showed that the diagonal airflowdesign resulted in a homogenous air distribution. The measuredspan of moisture content among the food buckets at a given dryingtime is tabulated in Table 2 in the form of standard deviation. Thehighest standard deviation was observed during constant rate ofdrying i.e. after four and sixth hours of drying time.

It was observed that at early stages of drying, first one or twobuckets in each set showed comparative a less value of moisturecontents. It can be observed at drying time of four, six and ninehours (Fig. 7) especially in last two sets of buckets. This wasbecause of not fully loading the dryer. As mentioned earlier thatdryer was not run at its full capacity but with a set of five bucketsat every two meter distance of drying chamber in a trial. Sidewalls of buckets were perforated, so some percentage of the aircoming through the neighbored empty bucket passed over tothe adjacent first bucket of a set under trial. It facilitated theevaporation of surface moisture of food lying in the first bucketof a set more than the following buckets of that set. Because sur-face moisture is loosely bounded so easy to evaporate. But eventhough the maximum variation (higher standard deviation) waslow (Table 2).

Material started to dry with same moisture contents and endedup at almost same level of moisture. Thus, the diagonal airflowdesign provided another way to lower down the problem of dryingheterogeneity in batch dryers.

5. Comparison between the experimental and simulated data

Simulation of velocity distribution showed turbulence when airentered in each bucket. So the measurements of velocity distribu-tion within a bucket at its different positions were necessary totake an average velocity of airflow. For this, in CFD solution, lineswere drawn passing through each bucket in longitudinal directionat three points (front, middle and back) with three positions (top,middle, bottom) for each point.

Fig. 6. Drying curves for five sets of trials at various distances in drying chamber (a) 1–5 buckets at 2 m distance (b) 6–10 buckets at 4 m distance (c) 11–15 buckets at 6 mdistance (d) 16–20 buckets at 8 m distance (e) 21–25 buckets at 10 m distance. It shows comparison of drying trend among food buckets lying at various distances in thedrying chamber. Each set of five buckets covered a distance of 2 m in the drying chamber. A spatial drying homogeneity has been depicted.

W. Amjad et al. / Journal of Food Engineering 144 (2015) 148–155 153

On the other hand, during experiments, air flow through eachbucket was measured by inserting TA5 hotwire anemometer atthree points for each bucket. These points were made by drillingholes at roof of drying chamber to insert the probe of anemometerat the place of each bucket. For a single bucket, five to eight veloc-ity readings were measured for each of its three points at variousdepths and then an average velocity was calculated for that bucket.

These results of average velocity were compared correspondingwith the results extracted from the simulation. The result of statis-tical analysis showed good correlation (correlation coefficient of87.09%) for airflow distribution between the average predictedand the average experimental measured velocity as illustrated inFig. 8. The simulated and actual air distribution with diagonalairflow design strengthen the design for uniform drying. This

Fig. 7. Comparative drying trend of food buckets at different time intervals of drying. It shows the drying trend for all the food buckets together during drying process.

Table 2Standard deviation among food buckets at different intervals of drying time.

Drying time (h) MC (kg/kg dry matter)

Median Minimum Maximum Standard deviation

2 2.89 2.71 2.99 0.074 1.94 1.67 2.18 0.136 1.23 1.10 1.51 0.129 0.55 0.33 0.76 0.11

12 0.24 0.14 0.36 0.0614 0.18 0.10 0.27 0.0415 0.13 0.09 0.21 0.03

Fig. 8. Comparison of experimental and predicted average air velocity throughbuckets.

154 W. Amjad et al. / Journal of Food Engineering 144 (2015) 148–155

comparison of air distribution not only support the uniformity ofdrying along the length of drying chamber (in each bucket) but alsoalong its height (all the trays in a single bucket). Practically it waschecked by measuring rate of moisture contents reduction for eachtray of a bucket and it was found almost uniform. So, uniformity

occurred not only along the length of dryer within food bucketsbut also found good along the height of each bucket. Therefore,chance of significant variation of drying in direction of depth ofdrying chamber has been reduced.

6. Conclusion

A new batch type dryer was developed with distinct feature ofdiagonal air flow design for uniform air distribution in dryingchamber. The design was modeled and analyzed using ANSYS-Flu-ent. Numerically and experimentally, a good distribution of air wasfound in the drying chamber which gave assertiveness to design. Itwas found that whole drying material (potatoes slices 4 mm thick,at 48 �C) lying in diagonally arranged buckets, dried with almostuniform drying rate. As all the food was being exposed to air atsame time so the remarkable usual fact of under and over dryinghas been reduced. It shows that this design would ultimately leadto quality drying because air distribution has significant influenceon product quality. Comparison between experimental and pre-dicted average values of air flow through buckets revealed goodcorrelation coefficients of 87.09%. It showed that newly designeddryer is able to distribute air satisfactorily uniform.

Further research is on the way to conduct for the optimizationof this new dryer with respect to quality parameters of the product.It needs to be optimized with the cost and quality for its matureimplementation.

Acknowledgements

This work is made possible through the help and support fromGerman Academic Exchange Service (DAAD) and INNOTECHDrying Technology, Stuttgart, Germany.

References

Adams, R.L., Thompson, J.F., 1985. Improving drying uniformity in concurrent flowtunnel dehydrators. Trans ASAE 28 (3), 890–892.

Amanlou, Y., Zomorodian, A., 2010. Applying CFD for designing a new fruit cabinetdryer. J. Food Eng. 101 (2010), 8–15.

Ayensu, A., Asiedu-Bondzie, 1986. Solar drying with convective self-flow and energystorage. Solar Wind Technol. 3 (4), 273–279.

Babalis, S.J., Paparicolaou, E., Belessiotis, V.G., 2005. Impact of alternating drying-airflow direction on the drying kinetics of agricultural products. In: 3rd IASME/WSEAS Int. Conf. on Heat Transfer, Thermal Engineering and Environment,Corfu, Greece, August 20–22, 2005, pp. 300–305.

Darabi, H., Zomorodin, A., Akbari, M.H., Lorestani, A.N., 2013. Design of cabinet dryerwith two geometric configurations using CFD. J. Food Sci. Technol., doi 10:1007/s13197-013-093-1.

W. Amjad et al. / Journal of Food Engineering 144 (2015) 148–155 155

Fluent 6.3 user’s Guide, 2005. Fluent documentation. Fluent Inc, Lebanon.Gül ah, Çakmak, Cengiz, Y.ld.z., 2009. Design of a new solar dryer system with

swirling flow for drying seeded grape. Int. Commun. Heat Mass Transfer 36 (9),984–990.

Jacek, S., Andrzej, J.N., Dawid, R., 2010. Improved 3-D temperature uniformity in alaboratory drying oven based on experimentally validated CFD computations. J.Food Eng. 97 (2010), 373–383.

Janjai, S., Intawee, P., Chaichoet, C., Mahayothee, B., Haewsungcharern, M., Muller, J.,2006. Improvement of the air flow and temperature distribution in aconventional longan dryer. In: International symposium towards sustainablelivelihoods & ecosystem in mountaneous regions, Chiang Mai, Thailand.

Kiranoudis, C.T., Karathanos, V.T., Markatos, N.C., 1999. Computational fluiddynamics of industrial batch dryers of fruits. Drying Technol. 17 (1&2), 1–25.

Krll, K., 1978. Trockner und Trocknungsverfahren, . second ed. Trocknungstechniksecond ed., vol. 2 Springer, Berlin.

Margaris, D.P., Ghiaus, A.G., 2006. Dried product quality improvement by air flowmanipulation in tray dryers. J. Food Eng. 75, 542–550.

Mathioulakis, E., Karathanos, V.T., Belessiotis, V.G., 1998. Simulation of airmovement in a dryer by computational fluid dynamics: application for thedrying of fruits. J. Food Eng. 36 (2), 183–200.

Mirade, P.S., 2003. Prediction of the air velocity field in modern meat dryers usingunsteady computational fluid dynamics (CFD) models. J. Food Eng. 60 (2003),41–48.

Misha, S., Mat, S., Ruslan, M.H., Sopian, K., Salleh, E., 2013. Review on the applicationof a tray dryer system for agricultural products. World Appl. Sci. J. 22 (3), 424–443.

Müller, J., Heindl, A., 2006. Drying of medicinal plants. In: Bogers, R.J., Craker, L.E.,Lange, D. (Eds.), Medicinal and Aromatic Plants: Agricultural, Commercial,Ecological, Legal, Pharmacological and Social Aspects. Springer, Dordrecht, pp.237e252.

Roman, F., Strahl-schäfer, V., Hensel, O., 2012. Improvement of air distribution in afixed-bed dryer using computational fluid dynamics. Bio Syst. Eng. 112 (2012),359–369.

Shawik, D., Tapash, D., Srinivasa Rao, P., Jain, R.K., 2001. Development of an airrecirculating tray dryer for high moisture biological materials. J. Food Eng. 50,223–227.

Tzempelikos, D.A., Vouros, A.P., Bardakas, A.V., Filio, A.E., Margaris, A.P., 2012.Analysis of air velocity distribution in a laboratory batch-type tray air dryer bycomputational fluid dynamics. Int. J. Math. Comput. Simulat. Issue 6 (5).