International Journal of Reconfigurable and Embedded Systems (IJRES)

Vol. 14, No. 1, March 2025, pp. 176~183

ISSN: 2089-4864, DOI: 10.11591/ijres.v14.i1.pp176-183  176

Design of agrivoltaic system with internet of things control for

chili fruit classification using the neural network method

Wanayumini1, Habib Satria2, Rika Rosnelly3

1
Department of Informatic Engineering, Faculty of Engineering, Universitas Asahan, Kisaran, Indonesia
2
Department of Electrical Engineering, Faculty of Engineering, Universitas Medan Area, Medan, Indonesia
3
Department of Computer Science, Faculty of Engineering and Computer Science, Universitas Potensi Utama, Medan, Indonesia

Article Info ABSTRACT

Article history: Agriculture is a leading sector in the economy as well as the most dominant
provider of employment for the Indonesian people. The fertile soil factor
Received Jul 29, 2024 allows various types of fruit to be grown, including chilies. However,
Revised Sep 8, 2024 complex problems make chili farmers have limitations in implementing
Accepted Sep 28, 2024 conventional farming systems. Therefore, the development of an agrivoltaic
system with internet of things (IoT) integrated sensors on chili plants can
help farmers more easily control, add vitamins, fertilizers, and provide plant
Keywords: nutrients that can be done automatically periodically based on a real-time
clock schedule. This system also operates using photovoltaic (PV) as a
Agrivoltaic pumping machine for water circulation. Other technologies such as mini
Chili fruit smart cameras are also being developed to monitor and take pictures of
Control system chilies which will later be converted using the graphical user interface (GUI)
Internet of things application for segmentation. The method used in this chili fruit
Neural network method classification uses an artificial neural network in classifying ripe, raw, and
rotten chilies. The classification results obtained an R value of 0.9, which
means it is close to a value of 1 in the suitability of the chili image.
Therefore, farmers will find it easier to sort the chilies that will be harvested.
This is an open access article under the CC BY-SA license.

Corresponding Author:
Habib Satria
Department of Electrical Engineering, Faculty of Engineering, Universitas Medan Area
Medan, North Sumatra, Indonesia
Email: [email protected]

1. INTRODUCTION
The development of a smart agricultural system integrated with renewable energy should be able to
be developed in Indonesia due to the geographical and climate conditions that support it [1], [2]. Fertile soil
and adequate solar energy should help farmers export agricultural products and raw materials to other
countries to support the achievement of farmers' welfare [3], [4]. However, the problem of inadequate
technology has hampered agricultural productivity and of course has also had an impact on the decreasing
number of young people pursuing this profession. Therefore, the development of systematic technology in
farming should be able to stimulate the young generation and farmers to rise and compete again so that it
becomes the choice of the young generation and farmers today in pursuing this agricultural work. Several
modern agricultural systems have been developed today, including the agrivoltaic plant system [5]–[8]. This
system operates by combining photovoltaic (PV) and agricultural technology systems on the same land [9],
[10]. To support the advantages of the agrivoltaic system, this technology has also developed a model of
nutrient-rich water media as a substitute for soil [11], [12]. The plants that will be implemented are chili
plants which will also be developed in conditions that are resistant to disease and extreme weather [13]–[15].
Then the chili is planted using agrivoltaic which of course operates on all systems and is integrated with the

Journal homepage: http://ijres.iaescore.com

Int J Reconfigurable & Embedded Syst ISSN: 2089-4864  177

sensors that have been planted. This agrivoltaic model is also equipped with the nutrient film technique
(NFT) technique which is usually found in hydroponic systems [16], [17]. This technique involves flowing a
thin stream of nutrient solution beneath the plant roots through an inclined channel or pipe, so that the roots
are only partially submerged in the solution, allowing air to continue to flow around the roots [18]. This NFT
system can be integrated with internet of things (IoT) technology to monitor and control plant conditions
automatically [19], [20]. The development of IoT technology in this agrivoltaic system is capable of
monitoring and controlling agricultural systems to become more sophisticated. The IoT system works based
on sensors connected to the NodeMCU ESP8266 so that it can collect data in real time and send it to a central
platform that can be accessed online [21]–[23]. In the nutrient and temperature monitoring system in
agrivoltaic technology, sensors such as pH sensors, DHT11 humidity sensors, and nutrient sensors are
installed around the plants to measure nutrient conditions, ambient temperature, and water acidity. The
collected data is then sent via a website for further analysis. Then, in order for this system to work
continuously, this agrivoltaic technology is developed using solar power, of course, allowing the agrivoltaic
technology system to operate independently and save energy [24]. The usefulness of implementing IoT-based
technology is inseparable from avoiding systems that die due to grid power outages. With this hybrid system
created, it will prevent blackouts and missed monitoring when the grid power goes out, so that technology
can be controlled within 24 hours. In addition, the use of solar power as an energy source in this monitoring
system has several advantages, the first of which is solar power as a reliable new renewable energy source. In
remote or isolated agricultural environments, access to grid power supplies may be limited. By using solar
panels to generate energy, the monitoring system can operate independently without relying on external
power supplies. This will certainly ensure the smooth operation of the system and continuous monitoring.
This agrivoltaic system previously farmers had limitations in caring for their plants by using this technology
is able to provide easy access so that it does not have to be controlled continuously. Real-time monitoring
with the IoT system allows for accurate and up-to-date monitoring and control of nutritional and temperature
conditions in chili plants [25]. This provides direct information to users, allowing chili farmers to
immediately take necessary actions to maintain optimal growing conditions. By utilizing IoT technology and
solar energy, this system is expected to be an efficient, automatic, and reliable solution in chili cultivation
with the NFT system. Then to make it easier to classify ripe, raw, and rotten chilies, a neural network method
was developed with the help of data collection based on data from a mini smart camera. This smart farming
complex model is also very easy to implement in housing without a large area of land.

2. METHOD
Chili plant growth monitoring can be monitored remotely and chili fruit care can be controlled
remotely using the IoT system. This IoT-based agrivoltaic chili plant automatic control system technology is
designed to make farming easier while producing high-quality chilies. This system will also make it easier for
agrivoltaic chili farmers to carry out maintenance only via mobile phones, and allow farmers to monitor and
control anytime and anywhere as long as there is a stable internet connection. Chili plants that are monitored
are carried out in real-time by sending image data using a mini smart camera module and several integrated
sensors. Image data will be stored in the application for monitoring plant conditions during the planting period.
This tool is also equipped with an ESP8266 microcontroller which functions to transfer data from sensors in
the IoT concept. The special advantage of agrivoltaic technology developed from this water circulation system
is its ability to accelerate the growth of chilies because the nutrients in the water tank can be maintained
properly thanks to sensor readings, so that plants do not lack nutrients. In addition, this tool is also flexible and
can be applied to other plants, including those planted in soil media. Then the artificial neural network method
is used in the classification of chili fruits such as ripe, raw and rotten chili fruit patterns so that it is easier for
farmers to sort. The integrated agrivoltaic design system of sensors and IoT is shown in Figure 1.

Figure 1. IoT integrated agrivoltaic system

Design of agrivoltaic system with internet of things control for chili fruit classification … (Wanayumini)
178  ISSN: 2089-4864

3. RESULTS AND DISCUSSION

Network configuration and connectivity of the IoT-based agrivoltaic plant nutrient and temperature
monitoring system require adequate network infrastructure. This is because reliable connectivity must be
available so that data collected through sensors can be sent to the website platform smoothly. The
development of networks such as cellular networks, Wi-Fi, and other wireless communication technologies
has expanded the scope of connectivity and enabled efficient and real-time data delivery. With solar power in
the research system, this has played an important role in supporting the operation of the IoT-based
monitoring system. In the nutrient and temperature monitoring system for chili plants, solar panels are used
to generate the energy needed to operate the sensors and other electronic devices. The following is the
performance graph of the output produced by PVs shown in Figure 2. Based on the INA219 sensor test on
agrivoltaic technology in the Figure 2(a) frequency output value of DC voltage PV measured by the INA219
sensor obtained a mean value of 13.17 with Std.DV of 113, Figure 2(b) frequency output value of DC current
PV measured by the INA219 sensor obtained a mean value of 1.20 with Std.DV of 116, and Figure 2(c)
frequency output value of DC power PV measured by the INA219 sensor obtained a mean value of 15.78
with Std.DV of 1.061. Then the value of the measurement results of the PV output connected to the lamp
load as ultraviolet (UV) lighting in agrivoltaic technology is shown in Table 1.

(a) (b)

(c)

Figure 2. Frequency distribution graph system of PV output values measured by INA219 sensor frequency
output value of (a) DC voltage PV, (b) DC current PV, and (c) DC power PV

Table 1. Solar panel measurement results

No Time Voltage (V) Ampere (A) Power (W)
1. 7:00 am 13.20 1.25 16.50
2. 9:00 am 13.30 1.30 17.29
3. 11:00 am 13.15 1.29 16.96
4. 1:00 pm 13.19 1.29 17.01
5. 3:00 pm 12.98 1.00 12.98

Int J Reconfigurable & Embedded Syst, Vol. 14, No. 1, March 2025: 176-183
Int J Reconfigurable & Embedded Syst ISSN: 2089-4864  179

Table 1 shows the results of maximum power measurements at around 1:00 pm with a power of
17.01 W with the aim of supplying an electrical energy source to the UV lamp and also as a source for the
microcontroller. Testing data transfer on NodeMCU to Blynk or vice versa using the NodeMCU ESP8266
microcontroller. This test functions as a data transmission and data receiver so that the data obtained has the
stability of sending data testing sensors used on the NodeMCU ESP8266 microcontroller. This table may be
used to show the efficiency or reliability of data transmission by the NodeMCU ESP8266 module. The
difference between the data sent and received can indicate several factors such as data loss, signal quality, or
the module's ability to transmit data as shown in Table 2.

Table 2. Testing the NodeMCU ESP8266

No NodeMCU ESP8266 Transmitted data (kb) Received data (bytes)
1. Experiment 1 1.0 20
2. Experiment 2 1.2 23
3. Experiment 3 1.3 25
4. Experiment 4 1.5 33
5. Experiment 5 1.7 39
6. Experiment 6 1.8 42

In Table 2, the transmitted data can be obtained from 1.0 kb to 1.8 kb, directly proportional to the
amount of data received, there is also an increase from 20 bytes to 42 bytes. Therefore, the measurement of
the NodeMCU ESP826 that has been tested obtained good efficiency values and also success in data
transmission. Temperature and humidity testing changes throughout the day in the agrivoltaic technology
environment. Environmental conditions greatly affect the growth of chili plants under solar panels, as well as
to manage and optimize agricultural and energy production. The monitoring system at the temperature
measured in degrees Celsius (°C) at each time. This temperature varies from the lowest 28.3 °C at 7:00 am to
the highest temperature reaching 31.6 °C at 1:00 pm, then slightly decreased at 3:00 pm. The level of air
humidity measured in percent also varies relatively from 75% in the morning (7:00 am) to 65% at noon 1:00
am, and then slightly increased in the afternoon (66%). Results of temperature and humidity testing in
agrivoltaic are shown in Table 3. In Table 3, the temperature and humidity measurements are at stable and
normal conditions. The sensor works well where the temperature increases and humidity decreases so that the
agrivoltaic technology is stable and can operate efficiently.

Table 3. Results of temperature and humidity testing in agrivoltaic

No Time Temperature (°C) Humidity (%) Condition
1. 7:00 am 28.3 75 Normal
2. 9:00 am 29.6 72 Normal
3. 11:00 am 30.5 68 Normal
4. 1:00 pm 31.6 65 Normal
5. 3:00 pm 31.3 66 Normal

3.1. Chili fruit classification testing based on graphical user interface

After the agrivoltaic system works well, then testing the application that has been built with
MATLAB software is carried out. This development is designed based on a graphical user interface (GUI)
model with the aim of analyzing the quality or type of chili fruit based on digital images. Testing is carried
out by determining the accuracy of testing ripe, raw, and rotten chili fruit. The GUI system built based on the
neural network method by testing ripe chili fruit, then testing raw chili fruit and finally testing rotten chili
fruit. The chili fruit samples tested based on real data in the field use a mini smart camera module that has
been integrated with the IoT system. The data obtained is stored in a database and then data processing is
carried out using MATLAB software. The GUI system built based on the neural network method is shown in
Figure 3.
The GUI application operating system which is the heart of this system consists of the first on the
open file button allows users to open a chili fruit image file that will be analyzed by the system. Then on the
segmentation button with this button system may be used to start the segmentation process, which is dividing
the image into smaller parts or separate objects, such as chili fruit for further analysis. Chili classification,
namely this system works the chili classification process which may involve recognizing the type or quality
of chili based on images of ripe chili fruit, raw chili fruit, and rotten chili fruit. The reset button is used to
reset or reset the application, delete data, or images that have been loaded, so that users can start a new
process. Then there are three areas used to display images or analysis results, such as original images,
Design of agrivoltaic system with internet of things control for chili fruit classification … (Wanayumini)
180  ISSN: 2089-4864

segmentation results, and classification results. Figure 3 is the entire GUI system built based on the neural
network method. Then Figure 3(a) chili fruit testing classification system based on GUI which will be
compressed using neural network method, Figure 3(b) is image data that when inputting ripe chili fruit will
display the classification of ripe chili fruit, in Figure 3(c) testing raw chili fruit whose data is obtained based
on the image of green chili fruit with raw category and the last Figure 3(d) is the testing of rotten chili fruit so
that farmers will later find it easy to sort chili fruit. The system that will be used uses a neural network
method whose entire data has been input into a computer program with 2 input neurons, 2 hidden layers, and
1 output neuron with 10 and 5 neurons respectively. This structure is used to process input data and produce
the desired output after going through several levels of processing.

(a) (b)

(c) (d)

Figure 3. GUI system built based on neural network method (a) chili fruit testing classification system
based on GUI, (b) ripe chili fruit testing, (c) raw chili fruit testing, and (d) rotten chili fruit testing

3.1.1. Neural network training performance on chili data

The processed data is then determined to measure and monitor the performance of the neural
network. There are three charts that you can monitor: training, validation, and testing. By doing this, the
performance of the neural network changes during the training process and helps decide when to stop training
to avoid overfitting. When the model fits too well with the training data and loses its ability to predict new
data well. The following neural network training performance data is shown in Figure 4.
The results of the neural network training performance measurement shown in Figure 4, can be
obtained the best validation performance graph value achieved in the 10 th epoch with a mean squared error
(MSE) value of around 0.00063967. In the blue graph value, there is a decrease which means that the lower
the MSE in the train, the better the learning from the training data. In the green graph value is used in reading
the value to evaluate the model can operate generalization well to new data where the performance in epoch
10 is the best performance on validation data before the start of overfitting.

Int J Reconfigurable & Embedded Syst, Vol. 14, No. 1, March 2025: 176-183
Int J Reconfigurable & Embedded Syst ISSN: 2089-4864  181

Figure 4. Neural network training performance

3.1.2. The results of the regression graph test on the neural network model on chili fruit data
The test results are in the form of a regression plot of the performance of the artificial neural
network model in predicting data during the training, validation, and testing processes. This diagram can be
used to evaluate how close the model's output predictions are to the actual target values (targets) shown in
Figure 5. The result data in Figure 5 shows the performance of the neural network method has a very good
performance model on the value of the training and validation data programmed. Then there are 4 target
graphs, namely the blue graph where the fit graph value with the R value reaches 0.9 which means it is close
to the value of 1, thus the model is very suitable for the training data. The green validation diagram shows the
suitability of the model with valid data. Next is the red target graph, which shows the relationship between
model predictions and test data targets, and finally the gray graph, which shows the overall suitability of the
model to all data (training, validation, and testing).

Figure 5. Results of regression graph testing on neural network models

4. CONCLUSION
The integration of agrivoltaic technology with the development of the IoT system will have an
impact on farmers, especially chili farmers, in modern farming. Farmers can control and add vitamins,
fertilizers, and water quality based on a control system that has been programmed with the IoT system. Plants
are then given nutrients and watered automatically and regularly according to a real-time clock schedule.
This system is also equipped with a PV energy source (PV-off grid), so that water circulation through the
pump machine is never interrupted. The development of a small smart camera model was also developed to

Design of agrivoltaic system with internet of things control for chili fruit classification … (Wanayumini)
182  ISSN: 2089-4864

monitor chilies in taking pictures. This image is then converted for segmentation using a GUI that has been
programmed with MATLAB software. The method used in this chili fruit classification uses an artificial
neural network in classifying ripe, raw, and rotten chilies. The system to be used uses an artificial neural
network method whose entire data has been input into a computer program with 2 input neurons, 2 hidden
layers and 1 output neuron consisting of 10 and 5 neurons respectively. This structure is used to process input
data and produce output with the desired accuracy after going through several levels of processing. The
classification results obtained an R value of 0.9, which means it is close to a value of 1 in the suitability of
the chili image. Thus, farmers will find it easier to sort the chilies to be harvested.

ACKNOWLEDGEMENTS
The author would like to thank Asahan University for the Internal Grant Scientific Paper Publication
Assistance Programbased on Guidebook No. 103/KPTS-UNA/YEAR 2024. We would also like to thank
Universitas Medan Area for supporting this research. And research collaboration with Universitas Potensi
Utama.

REFERENCES
[1] A. E. Mulyono, P. Apnitami, I. S. Wangi, K. N. P. Wicaksono, and C. Apriono, “The potential of smart farming IoT
implementation for coffee farming in Indonesia: a systematic review,” Green Intelligent Systems and Applications, vol. 2, no. 2,
2022, doi: 10.53623/gisa.v2i2.95.
[2] H. Satria, S. Nisworo, J. Windarta, and R. B. Y. Syah, “Performance of single axis tracker technology and automatic battery
monitoring in solar hybrid systems,” Bulletin of Electrical Engineering and Informatics (BEEI), vol. 12, no. 6, pp. 3247–3255,
2023, doi: 10.11591/eei.v12i6.5506.
[3] Al-Khowarizmi, A. R. Lubis, M. Lubis, and R. F. Rahmat, “Information technology based smart farming model development in
agriculture land,” IAES International Journal of Artificial Intelligence (IJAI), vol. 11, no. 2, pp. 564–571, Jun. 2022, doi:
10.11591/ijai.v11.i2.pp564-571.
[4] A. Sambas, M. Mujiarto, G. Gundara, G. Refiadi, N. S. Mulyati, and I. M. Sulaiman, “Development of smart farming technology
on ginger plants in Padamulya Ciamis Village, West Java, Indonesia,” International Journal of Research in Community Services,
vol. 4, no. 3, pp. 93–99, 2023, doi: 10.46336/ijrcs.v4i3.483.
[5] S. H. Chae, H. J. Kim, H. W. Moon, Y. H. Kim, and K. M. Ku, “Agrivoltaic systems enhance farmers’ profits through broccoli
visual quality and electricity production without dramatic changes in yield, antioxidant capacity, and glucosinolates,” Agronomy,
vol. 12, no. 6, p. 1415, 2022, doi: 10.3390/agronomy12061415.
[6] U. Jamil, A. Bonnington, and J. M. Pearce, “The agrivoltaic potential of Canada,” Sustainability, vol. 15, no. 4, p. 3228, 2023,
doi: 10.3390/su15043228.
[7] P. E. Campana et al., “Experimental results, integrated model validation, and economic aspects of agrivoltaic systems at northern
latitudes,” Journal of Cleaner Production, vol. 437, 2024, doi: 10.1016/j.jclepro.2023.140235.
[8] A. S. Pascaris, R. Handler, C. Schelly, and J. M. Pearce, “Life cycle assessment of pasture-based agrivoltaic systems: emissions
and energy use of integrated rabbit production,” Cleaner and Responsible Consumption, vol. 3, 2021, doi:
10.1016/j.clrc.2021.100030.
[9] T. Krexner, A. Bauer, A. Gronauer, C. Mikovits, J. Schmidt, and I. Kral, “Environmental life cycle assessment of a stilted and
vertical bifacial crop-based agrivoltaic multi land-use system and comparison with a mono land-use of agricultural land,”
Renewable and Sustainable Energy Reviews, vol. 196, 2024, doi: 10.1016/j.rser.2024.114321.
[10] K. A. K. Niazi and M. Victoria, “Comparative analysis of photovoltaic configurations for agrivoltaic systems in Europe,”
Progress in Photovoltaics: Research and Applications, vol. 31, no. 11, pp. 1101–1113, 2023, doi: 10.1002/pip.3727.
[11] R. A. Gonocruz et al., “Analysis of the rice yield under an agrivoltaic system: a case study in Japan,” Environments, vol. 8, no. 7,
p. 65, 2021, doi: 10.3390/environments8070065.
[12] F. Johansson, B. E. Gustafsson, B. Stridh, and P. E. Campana, “3D-thermal modelling of a bifacial agrivoltaic system: a
photovoltaic module perspective,” Energy Nexus, vol. 5, 2022, doi: 10.1016/j.nexus.2022.100052.
[13] M. S. M. Pandi et al., “IoT based greenhouse condition monitoring system for chili plant growth,” Journal of Advanced Research
in Applied Sciences and Engineering Technology, vol. 41, no. 1, pp. 142–153, Nov. 2024, doi: 10.37934/araset.41.1.142153.
[14] A. A. A. Halim, R. Mohamad, F. Y. A. Rahman, H. Harun, and N. M. Anas, “IoT based smart irrigation, control, and monitoring
system for chilli plants using NodeMCU-ESP8266,” Bulletin of Electrical Engineering and Informatics (BEEI), vol. 12, no. 5, pp.
3053–3060, 2023, doi: 10.11591/eei.v12i5.5266.
[15] N. S. Abu, W. M. Bukhari, C. H. Ong, M. N. Sukhaimie, S. Wibowo, and A. M. Kassim, “Automated agricultural management
systems using smart-based technology,” International Journal of Emerging Technology and Advanced Engineering, vol. 12, no. 5,
pp. 123–131, 2022, doi: 10.46338/ijetae0522_14.
[16] R. S. N. A. R. Aris, K. I. F. Mohammad, L. S. Syafie, F. H. Azimi, and S. R. Sw, “Front-end development of nutrient film
technique for hydroponic plant with IoT monitoring system,” International Journal of Advanced Trends in Computer Science and
Engineering, vol. 9, no. 1.3, pp. 9–14, 2020, doi: 10.30534/ijatcse/2020/0291.32020.
[17] A. Farooq, A. K. Verma, C. M. Hittinahalli, N. Harika, and M. Pai, “Iron supplementation in aquaculture wastewater and its effect
on the growth of spinach and pangasius in nutrient film technique based aquaponics,” Agricultural Water Management, vol. 277,
2023, doi: 10.1016/j.agwat.2022.108126.
[18] D. Adidrana et al., “Simultaneous hydroponic nutrient control automation system based on internet of things,” International
Journal on Informatics Visualization, vol. 6, no. 1, pp. 124–129, 2022, doi: 10.30630/joiv.6.1.865.
[19] I. Agustian, B. I. Prayoga, H. Santosa, N. Daratha, and R. Faurina, “NFT hydroponic control using mamdani fuzzy inference
system,” Journal of Robotics and Control (JRC), vol. 3, no. 3, pp. 374–383, May 2022, doi: 10.18196/jrc.v3i3.14714.
[20] B. Siswanto, Y. Dani, D. Morika, and B. Mardiyana, “A simple dataset of water quality on aquaponic fish ponds based on an
internet of things measurement device,” Data in Brief, vol. 48, 2023, doi: 10.1016/j.dib.2023.109248.

Int J Reconfigurable & Embedded Syst, Vol. 14, No. 1, March 2025: 176-183
Int J Reconfigurable & Embedded Syst ISSN: 2089-4864  183

[21] D. Huo, A. W. Malik, S. D. Ravana, A. U. Rahman, and I. Ahmedy, “Mapping smart farming: addressing agricultural challenges
in data-driven era,” Renewable and Sustainable Energy Reviews, vol. 189. 2024. doi: 10.1016/j.rser.2023.113858.
[22] S. Pal, H. VijayKumar, D. Akila, N. Z. Jhanjhi, O. A. Darwish, and F. Amsaad, “Information-centric IoT-based smart farming
with dynamic data optimization,” Computers, Materials and Continua, vol. 74, no. 2, pp. 3865–3880, 2023, doi:
10.32604/cmc.2023.029038.
[23] A. Piancharoenwong and Y. F. Badir, “IoT smart farming adoption intention under climate change: the gain and loss perspective,”
Technological Forecasting and Social Change, vol. 200, 2024, doi: 10.1016/j.techfore.2023.123192.
[24] H. Satria et al., Potential microgrid model based on hybrid photovoltaic/wind turbine/generator in the coastal area of North
Sumatra,” Indonesian Journal of Electrical Engineering and Computer Science (IJEECS), vol. 34, no. 2, pp. 768–776, May 2024,
doi: 10.11591/ijeecs.v34.i2.pp768-776.
[25] H. Satria, I. Dayana, R. B. Y. Syah, D. Noviandri, M. K. Zuhanda, and R. Salam, “Earthquake detection in mountainous homes
using the internet of things connected to photovoltaic energy supply,” International Journal of Reconfigurable and Embedded
Systems (IJRES), vol. 13, no. 2, pp. 315–322, Jul. 2024, doi: 10.11591/ijres.v13.i2.pp315-322.

BIOGRAPHIES OF AUTHORS

Wanayumini received a S.Kom. degree in information management or software

with expertise in system analyst at STMIK YPTK Padang, West Sumatra in 1995, and
M.Kom. degree in information technology in artificial intelligence (AI) from UPI YPTK
Padang, West Sumatra, in 2008 and in 2021 successfully obtained a Dr. degree in computer
science from the University of North Sumatra, in the field of expertise in data mining, image
processing, machine learning, and artificial intelligence. Currently a lecturer at the
Department of Informatics Engineering, Universitas Asahan, Indonesia. Her research fields
are data mining, image processing, machine learning, artificial intelligence, and decision
support system. She can be contacted at email: [email protected].

Habib Satria received B.Sc. degree in electrical engineering education from

Padang State University in 2016, and M.T. degree in electrical engineering from Universitas
Andalas, Indonesia, in 2018 and engineer professional (Ir.) degree from Universitas
Diponegoro, Indonesia, in 2022. He is currently a lecture in Department of Electrical
Engineering, Universitas Medan Area, Indonesia. His research interests are new and
renewable energy, concerning about solar power plant, automatic control system, IoT, real-
time simulation, green computing, pattern recognition, and power system. He is a member of
the International Association of Engineers (IAENG) and The Institution of Engineers
Indonesia. He can be contacted at email: [email protected].

Rika Rosnelly received S.Kom. degree in computer science STMIK Logika

Medan in 2005, and M.Kom. degree in computer science from Putera Indonesia University
YPTK Padang, Indonesia, in 2008 and Dr. degree from Gadjah Mada University Yogyakarta
in 2017. Currently she is a Professor at Department of Computer Science, Faculty of
Engineering and Computer Science, Universitas Potensi Utama, Indonesia. Her research
lines are image processing, artificial intelligence, and decision support system. She can be
contacted at email: [email protected]; [email protected].

Design of agrivoltaic system with internet of things control for chili fruit classification … (Wanayumini)