p–ISSN: 2723 – 6609 e-ISSN: 2745-5254
Vol. 5, No. 12, December 2024 http://jist.publikasiindonesia.id/
Indonesian Journal of Social Technology, Vol. 5, No. 12, December 2024 5883
Application of Fuzzy Logic for Android-Based Tempeh
Fermentation Process Control and Monitoring
Isac Ilham Akbar Habibi1*, Septriandi Wirayoga2, Miftakhul Huda3, Guntur
Yanuar Astono4
Universitas Negeri Malang, Indonesia
Email: [email protected]
*Correspondence
ABSTRACT
Keywords: tempeh
fermentation, fuzzy logic,
android application, MQ –
137, VNH2SP30.
Tempeh is a highly nutritious food source that is widely
consumed by people every day. Tempeh artisans, such as
those in Jatisari Village, produce quality tempeh through an
optimal fermentation process. The ideal fermentation
process takes place at a temperature of 30°C–35°C, a
humidity of 65%–72%, and an ammonia gas content of 43–
50 ppm, with a fermentation time of about 18 hours to
produce cooked tempeh. To improve the efficiency and
quality of the fermentation process, a tool that is able to
control and monitor the condition of the fermentation
chamber is needed automatically. This research develops a
tempeh fermentation control and monitoring system using a
NodeMCU microcontroller, relay, VNH2SP30, DHT11
sensor, and MQ-137 sensor, which are integrated with fuzzy
logic. This system controls the incandescent lamp as a heater
and the peltier as a cooler in the fermentation container
compartment. Control and monitoring results are displayed
in real-time via LCD and Android app, with the successful
fermentation time shortened to 16 hours.
Introduction
Tempeh is one of the highly nutritious food sources that is widely consumed
because of its affordable price. Tempeh is made from the main ingredient in the form of
soybean seeds, equipped with ingredients to support the fermentation process, such as
Rhizopus oligosporus mold, Rh. oryzae, Rh. stolonifer (bread mold), or Rh. Arrhizus
(Djunaidi, Purwanto, Ningrum, Jatnika, & Kabidoyo, 2020)Tapai Ripeness Monitoring
Application Using Fuzzy Tahani Method. The fermentation process is an important part
of producing quality tempeh, and many tempeh artisans, including in Jatisari Village,
strive to produce good products for consumption (Faqih & Miharja, 2024). The tempeh
fermentation process is one of the important stages in tempeh making that determines the
final quality of the product. This process requires careful control of temperature,
humidity, and ammonia gas levels in order to produce high-quality tempeh. In general,
fermentation of tempeh takes about 18 hours to reach a state ready for consumption.
However, the success of fermentation is greatly influenced by the stability of these
Isac Ilham Akbar Habibi, Septriandi Wirayoga, Miftakhul Huda, Guntur Yanuar Astono
Indonesian Journal of Social Technology, Vol. 5, No. 12, December 2024 5884
environmental parameters (Arifwidodo, Oktavian, & Ginting, 2022). The optimum
temperature for fermentation of tempeh ranges from 30°C to 35°C, with an ideal humidity
level between 65% to 72%. In addition, the level of ammonia gas produced during the
fermentation process must be in the range of 43 to 50 ppm. Inaccuracies in controlling
these parameters can cause fermentation to run suboptimally, thus affecting the quality,
texture, taste, and nutritional value of tempeh (Mahali, 2016).
The need for equipment capable of automatically controlling and monitoring the
fermentation process is very relevant. This kind of device not only helps to keep the
fermentation environmental parameters within the ideal range, but also improves the
efficiency of the tempeh production process. In this context, the use of fuzzy logic can be
one of the innovative solutions (Said, Damanik, Yusri, Sofi, & Purnomo, 2024). Fuzzy
logic is an intelligent control method that is able to handle variability and uncertainty in
the system, so it is suitable for being applied to controlling temperature, humidity, and
ammonia gas levels in the tempeh fermentation room (Purwanto et al., 2023). With fuzzy
logic, the system can make automatic adjustments to changes in environmental conditions
in real-time, ensuring that fermentation parameters remain stable and in accordance with
requirements.
The implementation of fuzzy logic in controlling the tempeh fermentation process
requires hardware consisting of a microcontroller and various supporting sensors.
Microcontrollers, such as NodeMCUs, are used as the brains of the system that manages
the data from the sensors and controls the actuators. Temperature and humidity sensors,
such as DHT11 or DHT22, are used to monitor environmental conditions in fermentation
chambers. Meanwhile, gas sensors, such as MQ-137, serve to detect the levels of
ammonia gas produced during the fermentation process (Valencia, Purnama, Tjong, &
Liman, 2022). The data obtained from these sensors will be processed by the
microcontroller using fuzzy logic algorithms to determine the actions that need to be
taken, such as turning the incandescent lamp on or off as a heater, as well as turning the
peltier on or off as a cooler (Abidin, Supriyanto, Surtono, & Suciyati, 2024).
In addition to hardware, software also plays an important role in this system. Fuzzy
logic algorithms are designed to process data from sensors and generate the right
decisions. For example, if the temperature in the fermentation chamber is too low, the
system will increase the heating intensity by regulating the power of the incandescent
lamp. Conversely, if the temperature is too high, the system will activate the peltier to
cool the room. Likewise with humidity, if it is detected too low, the system can activate
the humidifier to increase the humidity until it reaches the desired set-point. In the case
of ammonia gas, if the gas level exceeds the specified threshold, the system can activate
ventilation to reduce the concentration of gas in the fermentation chamber.
This control and monitoring system is designed to be integrated with an Android
application to make it easier for users to monitor the fermentation process in real-time.
Data from the sensor is transmitted to the application via a wireless connection, such as
Wi-Fi, which is facilitated by the NodeMCU microcontroller. The Android app displays
information about temperature, humidity, and ammonia gas levels in an easy-to-
Application of Fuzzy Logic for Android-Based Tempeh Fermentation Process Control and
Monitoring
Indonesian Journal of Social Technology, Vol. 5, No. 12, December 2024 5885
understand graphical form, so users can monitor fermentation conditions remotely. In
addition, the app can also provide notifications in case of abnormal conditions, such as
excessive temperatures or ammonia gas levels that exceed safe limits, so users can
immediately take the necessary action.
The advantage of this system lies not only in its ability to maintain the stability of
fermentation parameters, but also in its efficiency in reducing fermentation time. Based
on the results of the research, this fuzzy logic-based control system is able to cut the
fermentation time from 18 hours to only 16 hours. This certainly provides significant
benefits for tempeh artisans, especially in terms of productivity and production cost
efficiency. With shorter fermentation times, artisans can increase production capacity
without sacrificing product quality.
Method
System Planning
Figure 1
Block Diagram of the Whole System
Figure 1 illustrates the overall system diagram block. There is a DHT11 sensor that
functions as a sensor for the temperature and humidity level of the room where the tempeh
fermentation process takes place and an MQ-137 sensor as a detector for ammonia gas
produced in the tempeh fermentation process. The readings from the two sensors will be
processed in the microcontroller.
The microcontroller will control the DC Motor Driver VNH2SP30 and relay
connected to the peltier thermoelectric cooler as well as the 5-watt lamp. The readings of
temperature, humidity and ammonia gas are displayed on the LCD and the android app.
Formation of the Fuzzy Set
In the process there are 5 conditions based on fuzzy logic that are created. This
fuzzy logic scheme has 3 types of member sets, namely 3 members of the input set
(Temperature, Humidity, NH3 Gas Content) and 2 members of the output set (Lamp,
Peltier).
Isac Ilham Akbar Habibi, Septriandi Wirayoga, Miftakhul Huda, Guntur Yanuar Astono
Indonesian Journal of Social Technology, Vol. 5, No. 12, December 2024 5886
Results and Discussion
1) Temperature Variable Membership Function
The first input of the temperature member set = {Cool, Normal, Hot} which has a
speaker universe {0, 50} where the parameter i {0, 25, 50}. The temperature variable
membership function for μCold {0, 5, 10, 15, 20, 25}, μNormal {20, 25, 30}, and for
μHot {25, 30, 35, 40, 45, 50} is as shown in Figure 2.
Figure 2
Temperature Variable Membership Function
Then for the interval range classification table for each set of temperature members
that has been made, it is shown in table 1.
TABLE 1
CLASSIFICATION OF TEMPERATURE MEMBER SET INTERVALS
Classification Interval
Cold 0°C – 20°C
Usual 21°C - 30°C
Hot >30°C
2) Fungi Membership Variable Humidity
The second input of the moisture member set = {Too Dry, Ideal, Too Humid} which
has a speaker universe {0 100} where the parameter i {0 50 100}. The humidity variable
membership function for μ Terlalu_Kering {0, 10, 20, 30, 40, 50, 55}, μIdeal {45, 55,
65}, and μ Terlalu_Lembab {55, 60, 70, 80, 90, 100} corresponds to Figure 3.
Figure 3
Humidity Variable Membership Function
Application of Fuzzy Logic for Android-Based Tempeh Fermentation Process Control and
Monitoring
Indonesian Journal of Social Technology, Vol. 5, No. 12, December 2024 5887
Then for the interval range classification table for each set of moisture members
that has been made, it is shown in table 2.
TABLE 2
CLASSIFICATION OF MOISTURE MEMBER SET INTERVAL
Classification Interval
Too Dry <45%
Ideal 45% - 65%
Too Moist >65%
3) Ammonia Gas Variable Membership Function
The third input of the member set Kadar_Gas_NH3 = {Belum_Matang, Mature,
Overmature} which has a speaker universe {0 100} where the parameter i {0 50 100}.
The humidity variable membership function for μ Belum_Matang {0, 10, 20, 30, 40, 43},
μRipe {45, 47.5, 50}, and μ Terlalu_Matang {51, 60, 70, 80, 90, 100} corresponds to
Figure 4.
Figure 4
Ammonia Gas Variable Membership Function
Then for the interval range classification table for each member set of NH3 gas
levels that have been made, it is shown in table 3.
TABLE 3
MEMBER SET INTERVAL CLASSIFICATION
GAS AMMONIA
Classification Interval
Immature <43 ppm
Ripe 44 ppm – 50 ppm
Overripe >51 ppm
4) Lamp Output Membership Function
After determining the membership function of each input variable, then determine
the membership function of the output that functions to run the actuator, namely a 5-watt
lamp and a peltier thermoelectric cooler (Guo, Li, & Li, 2019). For the first output
member function, i.e. the lamp shown in figure 5.
Isac Ilham Akbar Habibi, Septriandi Wirayoga, Miftakhul Huda, Guntur Yanuar Astono
Indonesian Journal of Social Technology, Vol. 5, No. 12, December 2024 5888
Figure 5. Fuzzy Logic Output Lamp Actuator
To achieve 2 output conditions as shown in Figure 5 must meet the following
conditions:
1. Output Off, the light turns off when the temperature is above the set-point, which is
36°C.
2. Output On, the light turns on when the temperature is below the set-point which is
<35°C.
5) Peltier Thermoelectric Cooler Output Membership Function
For the first output member function, i.e. the lamp shown in figure 6.
Figure 6 Fuzzy Logic Output Peltier Actuator
To achieve the 3 output conditions as shown in Figure 3.9 must meet the following
conditions:
1. Level 1 output, Driver VNH2SP30 generates a PWM signal of 0 PWM indicating the
peltier is in the off position.
2. Level 2 output, Driver VNH2SP30 generates a PWM signal of 128 PWM signaling a
live peltier with a Vout of 6V.
3. Level 3 Output, Driver VNH2SP30 generates a PWM signal of 255 PWM indicating
the peltier is in the live position with a Vout VNH2SP30 of 12V.
6) Rule Base
After the membership function is formed, each input and output variable then
creates a rule base. There are 18 rules that apply to decision-making in this system
according to Figure 7.
Application of Fuzzy Logic for Android-Based Tempeh Fermentation Process Control and
Monitoring
Indonesian Journal of Social Technology, Vol. 5, No. 12, December 2024 5889
Figure 7 Rule Fuzzy Logic
The last one is a rule viewer where we can enter the input value and see the output
produced.
Figure 8 Rule Fuzzy Logic
Sensor Node Assembly
All existing components will be connected to the NodeMCU as a microcontroller.
It will be assembled according to Figure 9. on the sensor node.
Figure 9 Overall Tool Sensor Node
Hardware Manufacturing Results
Figure 10 is the final result of the pre-designed hardware assembly.
Isac Ilham Akbar Habibi, Septriandi Wirayoga, Miftakhul Huda, Guntur Yanuar Astono
Indonesian Journal of Social Technology, Vol. 5, No. 12, December 2024 5890
Figure 10
Tempeh Fermentation Box Seen from the Inside
Figure 11
Tempeh Fermentation Box Top View
Software Creation Results
In this section, we explain the parts in the design of the application display on an
android smartphone for the tempeh fermentation control system. The design of this android
application is carried out using the 7.0 software framework.
Figure 11
App Main View
DHT11 Sensor Testing
The test of the DHTll sensor aims to find out how accurate the data results read by the
DHTl1 sensor against the Hygrometer measuring instrument. Later, the output value of
DHTll will affect the condition of the tempeh fermentation container box.
Application of Fuzzy Logic for Android-Based Tempeh Fermentation Process Control and
Monitoring
Indonesian Journal of Social Technology, Vol. 5, No. 12, December 2024 5891
Figure 12
DHT11 Sensor Value Accuracy Testing
Based on the testing of the DHT11 sensor, the data that has been obtained is
presented in a table as shown in the following table IV:
TABLE 4
DHT11 SENSOR ACCURACY DATA
Testing Sensor
Value (%)
Higrometer
Value (%)
Error
(%)
1 49% 47% 4.2
2 48% 47% 2.1
3 48% 47% 2.1
4 48% 47% 2.1
5 48% 47% 2.1
6 48% 47% 2.1
7 48% 47% 2.1
8 49% 47% 4.2
9 49% 47% 4.2
10 49% 47% 4.2
Average 2.94
MQ-137 Sensor Testing
The MQ137 sensor test aims to find out how accurate the data results read by the
MQ137 sensor against the ammonia meter measuring device. Later, the output value of
MQ-137 will affect the condition of the tempeh fermentation container box.
Figure 13
MQ-137 Sensor Accuracy Testing
Isac Ilham Akbar Habibi, Septriandi Wirayoga, Miftakhul Huda, Guntur Yanuar Astono
Indonesian Journal of Social Technology, Vol. 5, No. 12, December 2024 5892
The data generated from the MQ-137 sensor accuracy test are presented in the form
of a table as follows:
TABLE 5
MQ-137 SENSOR ACCURACY DATA
Testing
To-
Test Results
Ammonia
Meter
(ppm)
MQ-137
(ppm)
Error
(%)
1 55.9 55.7 0.44
2 56.4 55.2 0.22
3 55.5 55.3 0.44
4 55.8 55.7 0.22
5 57.2 57.3 0.21
6 55.2 55.4 0.44
7 56.2 56.1 0.22
8 61.5 61.7 0.45
9 60.6 60.5 0.2
10 56.7 56.8 0.21
Average 0.31
Comparison of Tempeh Fermentation Test Results with Tools and Without Tools
Then after the test data was taken. Furthermore, the duration of tempeh fermentation
is compared between testing with equipment with conventional or traditional tempeh
fermentation. The results of the comparison are shown in table 6 as follows:
TABLE VI
COMPARATIVE DATA ON THE FERMENTATION PROCESS OF TEMPEH WITH
AND WITHOUT TOOLS
NO
With Tools No Tools
Hour Information Hour Information
1
06 .00
un t i l
10 .00
a .m.
Raw
06 .00
to
06 .00
10 .00
Raw
2
11 .00
un t i l
20 .00
Par t ia l ly
Grow
Mold
11 .00
un t i l
10 :00
p .m.
Par t ia l ly
Grow
Mold
3
21 .00
un t i l
10 :00
p .m.
Ripe
10 :00
p .m.
00 .00
Ripe
In table 6, the difference between fermentation results with equipment with
fermentation without using equipment or traditionally. The tempeh fermentation process
uses tools from 06.00 WIB to 22.00 WIB while the fermentation process without using
Application of Fuzzy Logic for Android-Based Tempeh Fermentation Process Control and
Monitoring
Indonesian Journal of Social Technology, Vol. 5, No. 12, December 2024 5893
tools is from 06.00 WIB to 00.00 WIB. So from research, the tempeh fermentation
process using tools is 2 hours faster than without using tools.
Conclusion
The Mamdani fuzzy logic-based tempeh fermentation control and monitoring
system integrated with the application has succeeded in increasing the efficiency of the
tempeh fermentation process. With a temperature and humidity sensor reading accuracy
of 97.06%, as well as an ammonia gas sensor of 99.31%, the system is able to maintain
fermentation parameters according to the specified set-point. Temperature, humidity, and
ammonia gas levels are regulated to produce a fermentation process that is two hours
faster than conventional methods. This system not only increases the fermentation speed,
but also maintains the quality of the resulting tempeh. Real-time monitoring of
fermentation conditions through the application provides convenience for users. With this
tool, the productivity of tempeh MSMEs can increase significantly. This technology
proves the potential of fuzzy logic in the control of complex processes. In addition, the
use of accurate sensors ensures the stability of the fermentation environment. This
innovation opens up opportunities for the adoption of similar technologies in various
other fermentation industries. This research makes a real contribution in supporting
MSMEs to improve their production efficiency and quality.
Isac Ilham Akbar Habibi, Septriandi Wirayoga, Miftakhul Huda, Guntur Yanuar Astono
Indonesian Journal of Social Technology, Vol. 5, No. 12, December 2024 5894
Bibliography
Abidin, Ramadhani, Supriyanto, Amir, Surtono, Arif, & Suciyati, Sri Wahyu. (2024).
LPG Safety Monitoring System Using Artificial Neural Network with Back
Propagation Method Based on The Internet of Things (IoT). Journal of Energy,
Material, and Instrumentation Technology, 5(3), 100–108.
Arifwidodo, Bongga, Oktavian, Donny Arief, & Ginting, Jafaruddin Gusti Amri. (2022).
The Performance Analysis of Hybrid SDN–IP Reactive Routing on ONOS
Controller in Tree Topologies. 2022 IEEE International Conference on
Communication, Networks and Satellite (COMNETSAT), 118–122. IEEE.
Djunaidi, Karina, Purwanto, Yudhy Setyo, Ningrum, Rahma Farah, Jatnika, Hendra, &
Kabidoyo, Wali Syahputro Cahoyo. (2020). Tapai Ripeness Monitoring
Application Using Fuzzy Tahani Method. Journal of Physics: Conference Series,
1477(5), 52019. IOP Publishing.
Faqih, Abu Rizal, & Miharja, Rediawan. (2024). Quality Management Analysis Using
Seven Tools Case Study of Tinumpuk Tempe Factory. ProBisnis: Jurnal
Manajemen, 15(5), 831–840.
Guo, Jian, Li, Zhaojun, & Li, Meiyan. (2019). A review on prognostics methods for
engineering systems. IEEE Transactions on Reliability, 69(3), 1110–1129.
Mahali, Muhammad Izzuddin. (2016). Smart door locks based on internet of things
concept with mobile backend as a service. Elinvo (Electronics, Informatics, and
Vocational Education), 1(3), 171–181.
Purwanto, Heri, Nugraha, Rikky Wisnu, Ferdiansyah, Fahmi Reza, Dewi, Deshinta
Arrova, Sofian, Rudy, & Rizaldy, Muhammad Faridh. (2023). Sustainable Smart
Home IoT to Open and Close the House Fence using a Scanning Method.
International Journal of Advanced Computer Science and Applications, 14(10).
Said, Ade Irawan, Damanik, Tania Sephia Azzahra, Yusri, Ardhini Ramadhani, Sofi,
Nurdiyana, & Purnomo, Hari. (2024). Design of an Iot Based Automatic Lawn
Vacuum & Leaf Bag Using Usability and Work Posture Method. SHS Web of
Conferences, 189, 1003. EDP Sciences.
Valencia, Venna, Purnama, Louis Putra, Tjong, Chandra, & Liman, Johansah. (2022).
Rancang Bangun Alat Pendeteksi Kebocoran Gas LPG Berbasis Internet of Things
Dengan Katup Regulator Otomatis. Techné: Jurnal Ilmiah Elektroteknika, 21(2),
225–242.