Historically, drying was used as a method to preserve food. Extraction of fluid
substances from material is known as drying by this means that water is removed
from solids to a certain level with different techniques (Barrett
et al., 2005). Therefore, drying is moisture migration from material
in a specific period. Drying system is the aspects that effect the drying procedure,
namely, moisture content of the material that is intended to be dried the heating
temperature (Ceylan et al., 2007).
In dehydration process there are two stages of moisture loss first the phase
that water is evaporated and then after the evaporated water will be extracted.
Hence, drying is important in chemical and food processing. Throughout the drying
procedure high energy levels are consumed this is because of removal of moisture
from the body this phenomena makes drying a high energy consuming procedure
(Ivanova and Andonov, 2001; Teeboonma
et al., 2002).
Fruits drying attributes is dependent on a number of factors like sorption
equilibrium, density and thermal properties. Design of any kind of heating process
required knowledge about the materials density and thermal attributes. Shrinkage
since throughout drying changes will take place in volume and internal porosity.
These changes will lead to modification in shape and size of the final product,
mass transfer and dielectric properties (Carsky, 2008;
Desmorieux et al., 2008).
In general, the air drying graphs are composed of two stages: the first phase
is the constant rate in which free water is evaporated this period is controlled
by the heat given to the material and mass transfer rates so principally the
boundary layer has the main responsibility for the transport mechanism (Barrett
et al., 2005). The second phase is termed as the falling rate, which
is a complicated phenomenon due to the fact that the controlling factor is the
transport resistance within the particles to de, dried. Subsequently; these
finding can led to the conclusion that both exterior and interior conditions
are influential on the drying procedure.
Increasing amounts of fresh fruit waste and consumer demand for dried fruits
made the manufacturers interested to produce dried fruit products (Kiranoudis
et al., 1997). Accordingly to reduce the moisture content there a
number of different types of dryers designed. Batch fruit tray dryer with hot
air flow has been widely used among other techniques for fruit dehydration (Das
et al., 2001). This technique is based on transportation of hot air
in between the trays that material rests on. Obviously, higher circulation rates
result in high rates of and mass transfer from the fruits body. Moreover, increase
in temperature will have a direct effect on the quality indicators, namely;
color, shape, texture and nutrient components (Das et
Nowadays, fuzzy base controlling systems has been increased for industrial
applications. The fuzzy controller causes to achieve higher quality products
with optimal energy consumption. In previous works decision making was performed
based on non-fuzzy controller. Dryer machine usually captures different locations
of cabin temperature as its inputs and produces suitable output signals to control
its modules such as heater and fan.
The present study gives an overview on applying fuzzy principles on dryers.
The fuzzy base monitoring basics were implemented on fruit dryer platform with
laboratory scale. Cabin is equipped with 4 temperature sensors which samples
are captured using microcontroller processor of controller. The captured samples
are converted to digital value by analog to digital converter module of microcontroller.
These values will be inputs parameter of fuzzy controller and it provides two
output signals for fan and heater. Fuzzy controller outputs are between 0 and
1 which a mapping function is required to change output values as controlling
MATERIALS AND METHODS
Drying process: Drying is an energy consuming process that results in
evaporation of the water of the fruit body and removal of the moisture content.
In hot air dryers the heat required for the drying process is supplied by hot
air which cycles through the dryer and contacts the product in our case fruit
(Lewicki, 2006; Wang et al.,
2007). The fruit dryer designed is a cabinet types drier and basically batches
type equipment. The trays are rectangular shape with the dimension of 40x40
cm. The fruit dryer was tunnel shape and constructed of stainless steel with
an area of 0.5x1x1.5 m. The dryer consists of a heating control unit which supplies
the drying energy by an electrical heater, an electrical fan that creates forced
flow of hot air through the trays, measuring sensors and the chamber which has
been discussed earlier. The operation can be controlled which this part will
be discussed further. The drying hot air was achieved by the electrical heating
and controlled by the heating controller unit. The air was heated to the desired
temperature then enters to the drying chamber. The temperature controller can
alter the air temperature. Furthermore, the air flow rate was varied by utilizing
the speed controller of the fans and moreover by adjusting dampers in the dryer
design. The air velocity was controlled at approximately 2 cm above the surface
of the trays. The drier proposed in this study is manufactured with sixteen
trays in each wagon as shown in Fig. 1. As dryer is energy
consuming equipment due to requiring electrical energy.
|| Architecture of platform dryer system
So this energy is diverse into heat which increases the air temperature in
contact with foodstuff and dries them. Energy consumption is the secondary consideration
compared with quality in most of the fruit dryer designs.
The change in moisture levels can be calculated by measuring the following
where, MCWB is the moisture content water basis, MCDB is the moisture content
dry basis, MW is the mass of water, MS in the sample mass
and MTS is the total solid mass.
The change in the moisture level can be calculated by measuring the total mass
weight and the weight of water present in the product. The following formula
indicates the relationship between the moisture content wet basis and dry basis:
General equations for heat and mass transfer are shown in Eq.
5 and 6. Moreover, the energy consumption is added in
where, Q is the rate of heat transfer by convection, hC is the heat
conductivity, A is the surface area and Tsur and T∞are
the surface and air film temperature respectively. The Tsur and T∞are
shown in Fig. 2. Since hC depends on the force
convection with increasing the fan air, the velocity will increase; this fact
will led to increase in hC. The conclusion result is increase in
the rate of heat transfer.
where, -mC is the rate of mass transfer and the reason for its negativity
is because of the moisture leaving the object, k is the coefficient of mass
transfer, A is the area, wf saturated vapor pressure of the film
and wa is the partial pressure of the air stream.
where, Q is the energy leaving and ë is the latent heat of evaporation
meaning that heat needed to heat 1 kg of sample.
Dryer platform: Platform of proposed controller is a cabinet dryer machine,
which equipped with four level sensors. Due to variety of temperature in cabin,
controller requires to capture temperature of whole cabin to make a suitable
decision. Figure 1 shows the designed cabinet dryer. Pars
Dryer Engineering Group co. in Iran developed the studied platform at 2008.
It is a convective dryer that was designed as a multipurpose batch dryer.
Developed dryer is controlled by a monitoring interface, which designed with
low-cost, low-power and reliable principles. It is a multi-purpose controller
that is able to have 16 input sensors. User can define several program using
main keyboard and LCD display of board. Fig. 3 illustrates
the architecture of main controller of dryer. In previous works (Javanmard
et al., 2009), this controller was utilized as the monitoring system
of batch tea dryer. Controller board is able to provide the output signals in
two types, binary signals that are utilized as alarm and on/off command and
Pulse-Width Modulation (PWM) to smooth control of instruments such as fan and
Fuzzy controller: Main controller of system is a microcontroller base
circuit which is able to convert measured values by temperature to digital level
to be used as fuzzy system inputs. The fuzzy control enables a mapping between
temperature values and behavior of system. The relative speed of fan is α
which is defined between 0 and 1. In zero fans will be turned off and 1 shows
the fan must be spinning by maximum speed. To achieve a smooth speed control
for fan, Pulse-Width Modulation (PWM) technique is utilized to drive main fan
of system (Valentine, 1998).
|| Surface of fruit and cabin air temperatures
|| Schematic of deployed processor for dryer machine
Second deciding will be power of heater, which is shown with α. The output
value for α will be in range of 0 and 1. So, the PWM approach is utilized
to controller heater temperature. Deployed microcontroller is able to produce
PWM signals with different individual channels. System inputs are shown with
T1-4 which are placed in different levels of cabinet (T1
= 5, T2 = 65, T3 = 125 and T4 = 185 cm). Figure
4 shows the fuzzy membership function of input temperature. Three temperature
points are selected . The membership of speed control (α) for main fan
is shown in Fig. 5 which defined speeds. The fuzzy controller
has another output decision for heater power (α) which is shown in Fig.
According to systems outputs which are speed of fan and power of heater, there
are two types of fuzzy rules. The defined fuzzy rules for speed controlling
of fan are listed as:
||Rule-1: IF (T1 is C) and (T4 is W) →
||Rule-2: IF (T1 is T) and (T4 is W) → MF
||Rule-3: IF (T1 is W) and (T4 is W) → TO
||Rule-4: IF (T1 is C) and (T4 is T) SF
||Rule-5: IF (T2 is C) and (T4 is W) FF
||Rule-6: IF (T2 is T) and (T4 is W) MF
||Rule-48: IF (T3 is T) and (T4 is W) SF
||Membership function for cabin temperature (C: Cool, T: Tepid
and W: Warm)
||Membership function for circulation fan speed (To: Turn-off,
SF: Slow fan, MF: Moderate fan and FF: Fast fan)
||Membership function for power control of heater (LO: Low power,
MO: Moderate and HI: High power)
In order to save the processor time, 48 effective rules are selected. We use
C programming language to provide microcontroller program. The next fuzzy decision
is the power of heater which is made based on inputs values.
The 18 fuzzy rules are defined for heater output as the following list:
||Rule-1: IF (T4 is C) HI
||Rule-2: IF (T4 is T) and (T3 is C) HI
||Rule-3: IF (T4 is T) and (T2 is T) HI
||Rule-4: IF (T4 is T) and (T1 is T) MO
||Rule-18: IF (T4 is W) and (T2 is W) LO
After input intensity estimation, fuzzy function makes its decision and provides
two output values between 0 and 1. Next function is the mapping function, which
estimate PWM duty-cycle value to control fan and heater. In addition, there
are several non-fuzzy decisions, which were defined, in critical behaviors.
To prevent high temperature in cabin, a fan that is controlled directly by microcontroller
starts to pump hot air of cabin to outside. There is another alarm signal which
shows the happened error in system.
RESULTS AND DISCUSSION
The purpose of this study is to reach a fixed temperature in the cabin meaning
that the lowest tray and the upper tray will have the same temperature. This
fact will lead to having better quality of the final dried fruit product. In
the previous work that was controlled with non-fuzzy monitoring, the temperature
of trays differed with each other. That caused in a non-homogenous product.
Figure 7 shows the captured temperatures with cabins
sensors, which were not changed smoothly due to on and off switching of heater
to reach in between the defined temperature, range. In addition, T1
that is placed in lowest part of cabin had lower temperature than upper sensor
that is T4. Figure 8 illustrates captured temperature
values of cabin in fuzzy controller system. As it is clear, after arriving temperature
to defined level the all sensors values were in similar temperature. In
addition, the fuzzy commands of heater and fan controller fixed temperature
of cabin smoothly.
Power consumption of fuzzy controller was tested during a drying task. At the
first time system reached to drying temperature and then using smooth controlling
of heater and fan could achieve a approximately fix temperature in cabin. Due
to less changing in cabin temperature, the quality of products will be in higher
quality. In addition, power consumption will be saved. After reaching drying
temperature, heater was not turned on with maximum power. The power consumption
of fuzzy system and non-fuzzy system is shown in Fig. 9. As
it is clear in results, the fuzzy system had less switching and lower power
of heater and fan, which caused to decrease the average power consumption of
||Captured temperatures of cabin during a drying task with non-fuzzy
||Captured temperatures of cabin during a drying task with fuzzy
The average power consumption of both systems is calculated with following
where, i is the current of system during a drying task (Td) and
v is the voltage of system which was about 220 V.
Therefore, the average current of system is the important parameter in calculation
of power consumption. The average current of non-fuzzy system was about 5.91
A and average current for fuzzy system was about 4.92 A.
In this study, a fuzzy base controller system for convective fruit dryer was
proposed. Input of fuzzy system was temperature of cabin and outputs were heater
power and speed of fan. A microcontroller base controller circuit was deployed
for implementing fuzzy system. Temperature values captured using analog to digital
converter unit of microcontroller. Output signals were produced using pulse-width
modulation technique to smooth control of fan and heater. The results of cabin
temperature show the amenability of using the fuzzy control in dryer machines.
The power consumption of system was also improved with fuzzy base decision and
low power consumption was achieved using proposed technique.