IOP Conference Series: Materials Science and Engineering

PAPER • OPEN ACCESS

Design of fuzzy logic controller for temperature control of small-scale

food storage
To cite this article: M Latif et al 2019 IOP Conf. Ser.: Mater. Sci. Eng. 602 012024

View the article online for updates and enhancements.

This content was downloaded from IP address 178.171.44.182 on 07/09/2019 at 02:14

Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012024 doi:10.1088/1757-899X/602/1/012024

Design of fuzzy logic controller for temperature control of

small-scale food storage

M Latif1, M Muharam1, Darmawan1, Darwison1, and R R Costa1

1
Department of Electrical Engineering, Faculty of Engineering, Universitas Andalas,
Padang, Indonesia

E-mail: [email protected]

Abstract. A Fuzzy Logic Controller (FLC) is developed to control the temperature of small-
scale food storage. The FLC has two components such as the algorithm part and the hardware
part. For the algorithm part, The FLC takes two input parameter such as the temperature and
the humidity of the storage and takes one output parameter such as the speed of the fan. The
membership function is designed as following hot, fair, cold, and too cold as for the
membership functions of the temperature parameter. For the humidity membership functions
are dry, dry enough, normal, less moist, and moist. As for the fan speed, the membership
functions are fast, less fast, fair, slow, and too slow. On the other hand, for the hardware, the
DHT11 senses the temperature and the humidity. For the cooling process, thermoelectric is
used to decrease the temperature. The results show that the FLC can be used to control the
temperature of the storage to achieve as the minimum temperature as 12.5oC.

1. Introduction
Fuzzy Logic Controller (FLC) is one of the method to control the engineering process or system that
has been studied by researcher [1-3] and is applied in many engineering field such as fan rotation
speed controller [4], controller of HVAC of electric vehicle [5], and robust controller in wireless
sensor network [6]. Also, the FLC can be used to control temperature [7, 8]. Moreover, temperature
regulation or control is very important to gain some advantages such as maintaining room conditions
for comfort [5] and maintaining the temperature of cold thermal storage [9, 10]. To get good
conditions and high performances of temperature control need some cost and some complex
constructions, especially, for large-scale cold thermal energy applications [8, 9, and 11].
However, for a small-scale cold thermal storage, the cost and the complex constructions need to be
reduced. For example, the cold thermal energy source of the large-scale storage that uses compressor
and coolant liquid should be replaced by low-cost technology such as thermoelectric cooling [12].
Thermoelectric cooling has some advantages such as no CFC and a small-shape material size. This
technology can replace from the vapor compression refrigerator to the thermoelectric cooling
refrigerator.
Therefore, by combining the abilities of FLC and the thermoelectric cooling, this paper aims to
design the FLC for temperature control of small-scale cold thermal storage with the thermoelectric
cooling as the actuator of the cooling source.

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012024 doi:10.1088/1757-899X/602/1/012024

2. Research Method
This section describes the algorithm of FLC, the hardware of FLC and as well as the design of the
small-scale cold thermal food storage.

2.1. Algorithm of FLC

Given Figure 1, it is well known that FLC has three steps in processing the input parameter these are:

2.1.1. Fuzzification. It is a process to change the input from the form of crisp becomes a fuzzy form
is usually presented in the form of a set of fuzzy-set with a function of their membership function.
Fuzzification input has two variables which are the temperature and humidity of the storage and
fuzzification output has one variable which is the speed of the fan. For temperature, it is set about four
kinds of membership function, i.e., hot, fair, cold, and very cold. For humidity, the membership
functions are dry, dry enough, normal, less moist, and moist. Then, for the speed of the fan, the
membership function is fast, less fast, fair, slow, and too slow. The membership function of
temperature, humidity, and fan speed can be seen in Figure 2, Figure 3, and Figure 4 respectively.

Figure 1. The architecture of FLC [3]

Figure 2. The membership function of temperature

2
Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012024 doi:10.1088/1757-899X/602/1/012024

Figure 3. The membership function of humidity

Figure 4. The membership function of the speed of the fan

2.1.2. Inference system. It is a reference to describe the relationship between input and output
variables which are processed and generated fuzzy shape. To explain the relationship between the
input and output are usually using "IF-THEN." To generate this IF-THEN rules is by combining the
variables of input and output. This can be summarized in Table 1. It results in about 20 rules. One of
the rules, for example, is if the temperature is hot and the humidity is dry then the speed of the fan is
fast. In an implementation, the FLC uses the Mamdani method with the mean of the maximum for
deciding the output of the rules.

Table 1. Rule-based inference system

Temperature Hot Fair Cold Too cold
Humidity
Dry Fast Less fast Slow Too slow
Less dry Less fast Fair Slow Slow
Normal Fair Fair Fair Fair
Less Moist Less fast Less fast Fair Slow
Moist Fast Less fast Slow Too slow

2.1.3. Defuzzification. It is the process of modifying the variable shape of fuzzy to be crisp data
which can be sent to another process such as computing in the microcontroller.

2.2. Hardware of FLC

On the designed system, see Fig. 6, there is a DHT11 that is used as a sensor for measuring the
temperature and humidity of the room. The data obtained are then processed by a microcontroller that
is the temperature, and the humidity is to be created fuzzy rules. Then the result is the decision of

3
Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012024 doi:10.1088/1757-899X/602/1/012024

fuzzy logic in the form of a PWM signal to be sent to the fan through a motor driver. So that the fan
speed is regulated based on PWM value that is obtained by FLC. The speed of fan causes air flow hit
thermoelectric cooling device such that the air to be cold. Therefore, the room will be cold.

Sensor Arduino Driver

Fan
DHT11 Uno PWM

Figure 5. Block Diagram of FLC’s Hardware

2.3 The small-scale cold thermal storage

Figure 7 shows the designed storage as the small-scale cold thermal storage. The storage has
dimensions about 60 cm x 60 cm x 40 cm. There are four thermoelectric cooling devices with an
attached fan for flowing air in the chamber.

Figure 6. The designed storage

3. Results and analysis

In this section, the results of the research are divided into two kinds of implementation of the FLC.
The first implementation is using Matlab Simulation which is conducted by the programming of the
FLC. The results are shown in Figure 8, Figure 9, and Table 2.
The second implementation of the FLC is conducting by translating and installing the fuzzy
inference system in C programming into the microcontroller. The microcontroller reads the input
variables such as temperature and humidity of the storage from sensor DT11 and gives the output
variable such as the speed of the fan to the motor driver. The result can be seen in Table 2.
As shown in Table 2, it can be compared between the Matlab simulation and the microcontroller
implementation in term of the duty cycle of PWM. The duty cycle of both implementation shows that
the difference is small. It is about 4.67% on average.

4
Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012024 doi:10.1088/1757-899X/602/1/012024

Figure 7. Matlab editor the FLC

Figure 8. Matlab viewer of the membership function

In general result, from Table 2, the time that takes the storage to achieve the storage temperature to
the minimum temperature of 12.4o C is about 3 hours from the initial condition of 28.4o C. Even
though the temperature of the lowest one is not equal but close to the target minimum temperature of
thermoelectric cooling device that is 10o C. This shows thermoelectric cooling device has
disadvantage of the amount of heat flux that can be dissipated is limited [12]. However, the

5
Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012024 doi:10.1088/1757-899X/602/1/012024

thermoelectric cooling device has some advantages such as non-CFC, no need compression system
and small-scale storage.

Table 2. Results of the FLC

Hmd Time PWM Micro- PWM Matlab PWM
No T(℃)
(%) (second) controller (%) (%) Difference (%)
1 28.4 95 0 94.12 94.51 0.42
2 24.6 95 540 91.37 93.73 2.58
3 23.8 95 1080 90.20 93.73 3.91
4 22.2 95 1620 88.24 93.33 5.78
5 21.4 95 2160 87.06 92.94 6.76
6 20.0 95 2700 87.06 83.92 3.6
7 20.4 95 3240 87.06 86.27 0.9
8 19.2 95 3780 85.88 80.78 5.94
9 18.4 95 4320 83.92 80.39 4.21
10 17.0 95 4860 80.39 80.39 0
11 17.2 95 5400 80.78 80.39 0.49
12 16.8 95 5940 80.78 80.39 0.49
13 15.4 95 6480 84.31 80.39 4.65
14 14.2 95 7020 76.47 69.02 9.74
15 13.0 95 7560 57.65 52.94 8.16
16 12.4 95 8100 56.47 52.94 6.25
17 13.2 95 8640 61.57 56.47 8.28
18 12.8 95 9180 56.86 52.94 6.9
19 13.0 95 9720 57.65 52.94 8.16
20 12.5 95 10260 56.47 52.94 6.25
Average of PWM Difference 4.67
| |
Note: PWM Difference =

4. Conclusion
The FLC for temperature control of small-scale cold thermal storage has been proposed. The FLC was
first is implemented in Matlab simulation, and then it was implemented in a microcontroller. The
result shows that both implementations are close to the resulted PWM. The FLC also can conduct the
output of the research in term of the minimum temperature of storage about 12.4o C.

Acknowledgments
Authors gratefully acknowledge for funding the research to Faculty of Engineering, the University of
Andalas by research contract No. 075/UN.16.09. D/PL/2018.

References
[1] Lee C C 1990 IEEE Transaction on Systems, Man, and Cybernetics 20(2) 404-418.
[2] Lee C C 1990 IEEE Transaction on Systems, Man, and Cybernetic 20(2) 419-435.
[3] Jang J S R, Sun C T, Mizutani E 1997 Neuro-Fuzzy and Soft Computing: A Computational
Approach to Learning and Machine Intelligence. International Edition. New Jersey: Prentice-Hall
International, Inc. 1997.
[4] Bonato J, Mrak Z, and Badurina M 2015 Scientific Journal of Maritime Research 29 56-64.
[5] Ibrahim B S K K et.al. 2012 Procedia Engineering 41 904-910.
[6] Kapitanova K, Son S H, and Kang K D 2012 Ad Hoc Network 10 709-722.
[7] Jana A K 2010 IEEE Transaction on Industrial Informatics 6(1) 25-35.

6
Conference on Innovation in Technology and Engineering Science IOP Publishing
IOP Conf. Series: Materials Science and Engineering 602 (2019) 012024 doi:10.1088/1757-899X/602/1/012024

[8] Zhiqiang G, Trautzsch T A, and Dawson J G 2002 IEEE Transaction on Industry Application
38(2) 414-424.
[9] Abdulazeez M, Salami M J E, and Tijani I B 2011 Intelligent Control for Automation of Yam
Storage System Using Fuzzy Logic Controller. 2011 Third International Conference on
Computational Intelligence, Modelling and Simulation. 2011.
[10] Oro E et al. 2012 Applied Energy 99 513-533.
[11] Lin P-W and Chan C L 2017 Thermal Energy Storage Analyses and Design. Elsevier Inc.
[12] Taylor R A and Solbrekken G L 2018 IEEE Transaction on Components and Packaging
Technologies 31(1) 23-31.