Abstract−Cucumber is one of the horticultural commodities that has good prospects for cultivation because cucumber plants can be marketed domestically and abroad. Soil condition and fertility a very important factors in increasing agricultural growth and production. The ideal data for a good soil pH for planting cucumbers is 6-7, for a soil temperature of 18-30ºC and humidity of 50-60%. Ignorance of farmers about the value and condition of the land can lead to poor production of cucumber plants.
Therefore the authors created a control and monitoring system to monitor soil conditions or growing media in cucumber plants.
In this system, there is a soil pH sensor, temperature sensor, soil moisture sensor, and automatic sprinkler for fertilizer when the soil pH value is less than the specified limit. This system also applies the Internet of Things concept for sending data on the Telkom IoT Platform platform. Based on the test results of testing the soil temperature sensor, it gets an average error value of 0.67% and an accuracy value of 99.33%. Testing the soil moisture sensor obtains an error value and accuracy of 4.80% and 95.20%, respectively. Whereas in testing the pH sensor which was calibrated using the linear regression method, it obtained an average error value of 1.69% and an accuracy of 98.31%.
Keywords: Cucumber; Soil Fertility; MQTT; Soil pH; IoT
1. INTRODUCTION
Cucumber is one of the horticultural commodity crops that has good prospects for cultivation because this cucumber plant is usually marketed domestically and abroad. According to data from the Indonesian Central Statistics Agency (BPS), cucumber production in Indonesia will reach 471,941 tons in 2021. The amount increased by 6.95% compared to the previous year (2020) which amounted to 441,286 tons. Cucumber vegetable production fell from 2011 to 2017. In those seven years, cucumber production fell by around 18.52% from 521.53 tonnes in 2011 and in 2017 to 424,917 tonnes. However, from 2018 to 2021 cucumber production experienced an increase again. Even so, this number cannot yet match production in 2011[1].
Cucumbers are usually exported to potential markets such as Malaysia, Singapore, and especially Japan, which is the target market for cucumber exports in pickled form[2]. The need for cucumbers tends to continue to increase along with population growth, increasing living standards, and public awareness of the importance of nutritional value. Soil condition and fertility is an important factor in increasing plant growth and yields. Fertilizer is one component that plays an important role in reducing the risk of production failure in plants[3][4][5].
Providing fertilizer at the right dose can maintain fertility in the soil. However, on the other hand, if fertilizer is applied over the appropriate dose, in the long term it can have a negative impact on soil fertility and the environment[6]. Cucumber plants can generally grow in all types of soil for agricultural land, but to obtain high production and good quality, cucumber plants require a soil temperature of between 18-30°C, soil moisture of 50- 60%, and a pH of 6-7[7].
Regarding the soil fertility factor, cucumber farmers' ignorance of soil conditions such as temperature, humidity, and especially the pH value due to excessive fertilization can cause a decrease in cucumber production.
To maintain soil quality on agricultural land and continue to increase cucumber production, an Internet of Things- based planting media control and monitoring system is needed[8]. In this system, a system will be created with several sensors to determine soil fertility parameters such as pH sensors, temperature sensors, and soil moisture sensors. Apart from that, a method was used to test the pH sensor, namely the simple linear regression method.
Internet of Things (IoT) technology is a device concept that can send data to the internet as a medium without requiring human intervention. In simple terms, humans do not need to directly control IoT objects/devices.
However, humans can control these objects remotely[9][10][11]. As shown in Figure 1, the IoT architecture consists of four aspects, the first aspect is devices consisting of IoT devices, such as sensors, 10 smart devices, and devices that can connect to the internet. This device can communicate with networks and services via communication protocols. The second aspect, namely the network, consists of a network of devices and technology that allows tools to communicate with the internet. Some of the network protocols used in IoT include WiFi[12], Bluetooth[13], ZigBee[14], and LoRaWAN[15]. The third aspect, namely the platform consists of software and services that function to manage data that has been collected by IoT devices and provide services such as device management, data management, data analysis, and connecting to the cloud[16]. The fourth aspect, namely application, consists of applications and services that are used to process and analyze data on IoT devices, thereby enabling users to monitor, control, and automate IoT devices[11][17].
Figure 1. Internet of Things Architecture[18]
Several studies such as research [19] aim to monitor pH levels in the soil. The test was carried out on rice fields in Sidorejo Village, Bantul, Yogyakarta, then the measurement results were saved on a micro SD and blynk server on a cellphone or laptop. In this research, the use of a soil pH sensor obtained good accuracy values.
Research [20] aims to create a monitoring system with an Arduino Uno microcontroller, soil pH sensor, soil moisture sensor, and water pump automation when the soil moisture value is below the specified limit. Apart from that, the test results of the data obtained are sent directly to the blynk application. In this research, there is no automated system for soil pH.
2. RESEARCH METHODOLOGY
2.1 Research Flow
This research was carried out in several stages, the first stage was library research for problem formulation, data collection, system design, realization of tool creation, and system testing using the Arduino IDE, and the last stage was the analysis and conclusion stage. The following is the flow of research carried out by the author, which can generally be depicted in Figure 2.
Figure 2. Research Flowchart
Based on Figure 2, the research flow consists of several stages. The first stage is the process of studying literature regarding cucumber plants, as well as several studies related to control and monitoring systems for soil moisture, pH, and temperature. With several references obtained, the next step is to compare several of these references to determine the system design that will be created.
In this research, several parameters will be tested, namely testing the entire system and testing soil fertility conditions such as temperature, pH, and soil moisture. The next stage is the system design stage which includes hardware and software design. After the design and creation of the system is complete, the next stage is to check the system. If there are no errors or errors in the system, then continue with data collection and install the system on the cucumber plants. However, if there are still errors or mistakes in the system, the system will be repaired and after that, the system will be checked again.
The analysis in this research is the result of data obtained from sensor testing, pump testing, and overall system testing. The final stage in this research flowchart is in the form of conclusions and suggestions from the research that has been carried out.
Figure 3. Hardware Design
Figure 3 is a block diagram of the entire system which includes all input processes, data processes, and output from the system design. In the input section, there are three sensors, namely a soil pH sensor, a soil moisture sensor, and a soil temperature sensor. The temperature sensor used in this research is a DS18B20 sensor which will read temperature values on the ground. The ideal temperature for cucumber plants is 18-30°C. Furthermore, the soil moisture sensor used in this research is a capacitive soil moisture sensor. The data results from the humidity sensor and soil temperature sensor will later only be sent to a laptop or smartphone so that farmers can monitor it.
Meanwhile, the last sensor, namely the pH sensor, will work to read the pH value of the soil. The ideal soil pH for cucumber plants is 6-7, but cucumbers are still tolerant of a pH of 5.5 -7.5 as the minimum and maximum limits.
Next, the data from the three sensors will be processed by a microcontroller, namely Arduino Uno. The data sent will be processed and displayed on the LCD screen, and sent to the IoT platform which can be viewed via a laptop or smartphone as output from the system via the internet network with an additional module in the form of the GSM SIM800L module. Apart from that, at the output, there is a water pump, where the water pump will be filled with NPK fertilizer solution to raise the soil pH. The pump will turn on automatically if the pH value is below the specified limit. The schematic series of this research can be seen in the Figure 4.
Figure 4. Hardware Schematic Design
Figure 4 shows a schematic circuit of the control and monitoring system for planting media for cucumber plants, where in the schematic circuit there are several components with their respective functions. The first component is a temperature sensor which is connected to the Arduino Uno atmega328 microcontroller via pin D2, the capacitive soil moisture sensor is connected to the Arduino Uno atmega328 via pin A0, and the pH sensor is connected to the Arduino Uno atmega328 via pin A1. The next component is the GSM SIM800l module which is used as an additional internet connection module which is connected to the Arduino atmega 328 via the TXD pin to D10 on the Arduino Uno atmega 328, and the RXD pin to D11 on the Arduino Uno atmega 328. Then on the LCD the Vss pin is connected to ground, the VCC pin is connected to the 5V pin, the SDA pin to pin A4 and the SCL to pin A5. The last component is a relay which is connected to the water pump and microcontroller. Table 3.1 is a table of input and output for the entire system being designed. Apart from that, there are also 12 V and 9 V adapters as a power supply for the Arduino Uno and as a voltage input for the water pump used. The 9 V power supply is connected to the VIN pin on the Arduino uno.There is a flowchart of the tool's overall work system flow which consists of initialization programming process steps in the Arduino IDE software until the data can be displayed on the Telkom IoT platform as in Figure 5.
Figure 5. Flowchart System
Based on Figure 5, the first process begins with the program initialization process on the Arduino which includes sensor initialization, as well as preparing and checking the internet connection. Next, the sensor will read the temperature, humidity, and pH values of the soil. If the soil pH value read by the sensor is <5.5, the pump containing the NPK fertilizer solution will turn on automatically to increase the soil pH. However, if the soil pH is above ≥ 5.5 then the pump will not turn on. Next, the reading values of the pH, humidity, and soil temperature sensors will be displayed on the LCD and sent to the laptop or smartphone.
3. RESULT AND DISCUSSION
3.1 Result of System Design
System design includes hardware design and software design. This system uses Telkom IoT as an IoT platform to store data which is the output of the designed system. The design results can be seen in Figure 5. Based on Figure 5, the system design results consist of a project box that contains an Arduino Uno microcontroller, a GSM SIM800L module that works as internet connectivity, an LM9265 stepdown module as a voltage reducer, and a solid-state relay that functions to control the water pump. Apart from that, on the outside of the box, there are also several components such as an LCD which is located on top of the box, a DS18B20 temperature sensor to detect soil temperature, a pH sensor to detect soil pH, a capacitive soil moisture sensor to detect soil moisture, a 220V adapter, a 9V power supply. , as well as a water pump. In this research, the system designed will measure the values of temperature, humidity, and soil pH and then these data will be sent to the IoT platform. When the pH value is less than 5.5, the system will turn on the water pump to distribute the NPK fertilizer solution.
Figure 6. Result of System Design 3.2 Soil pH Sensor Testing
In testing the pH sensor, the author will test the level of accuracy, error, and precision of the soil pH sensor used in the research. Testing the pH sensor is carried out by comparing the pH value issued by the pH sensor and the
5 7 43
After obtaining a sample of the ADC value and PH value, the soil pH sensor was then calculated using the linear regression method to determine the relationship between the ADC value issued by the sensor (X) and the measured pH value (Y).
Figure 7. ADC and PH Linear Regression Graph
The sample results of the ADC values issued by the sensor and the measured pH values are made into a graph as in Figure 7. From this graph it can be concluded that the smaller the measured pH value, the greater the ADC value, and conversely, the greater the measured pH value, the smaller the ADC value issued by the sensor.
The distance from the ADC value issued by the sensor at each pH value is almost always stable, so the linear regression method can be implemented on the pH sensor to predict the pH value that will be read by the sensor.
Next, the linear regression results were tested using 3 soil samples with PH values of 5.5, 6.5, and 7 with 30 samples testing each sample. Test accuracy results can be seen in Table 2
Table 2. Soil PH Testing Measurement PH Avg PH
Meter
Avg PH Sensor
Avg Difference
Avg Error (%)
Avg Accuracy (%)
Avg Precision (%)
5,5 5,5 5,697 0,279 3,58 96,42 97
6,5 6,5 6,419 0,081333 1,3 98,7 99
7 7 6,989 0,010667 0,2 99,8 99
Based on Table 2, it is known that the results of the 3 soil PH samples have measurement results with an accuracy of above 96% and an error below 4%. Based on measurement standards, it can be stated that the results are quite good plus the precision results are above 97% with 30 testing samples.
3.3 Soil Temperature Sensor Testing
In this research, the soil temperature sensor used is the Dallas DS18B20 sensor. Testing of this sensor was carried out in three different conditions, namely at temperatures of 30°, 25°, and 13° Celsius, where 30 data samples were taken for each condition. The author will look for the error, accuracy, and precision values of the DS18B20 sensor.
The results of soil temperature testing can be seen in Table 3
Table 3. Soil Temperature Testing Measurement
Temperature (oC)
AVG Thermometer
(oC)
Avg Temperature
Sensor (oC)
Avg Difference
(oC)
Avg Error
(%)
Avg Accuracy
(%)
Avg Precision
(%)
13 13 12,92 0,203 1,56 98,44 98
25 25 24,998 0,038 0,15 99,85 99
Temperature (oC)
AVG Thermometer
(oC)
Avg Temperature
Sensor (oC)
Avg Difference
(oC)
Avg Error
(%)
Avg Accuracy
(%)
Avg Precision
(%)
30 30 29,908 0,092 0,31 99,69 99
Based on Table 3, it is known that the results of the 3 soil temperature samples have measurement results with an accuracy of above 98% and an error below 2%. Based on measurement standards, it can be stated that the results are quite good plus the precision results are above 98% with 30 testing samples.
3.4 Soil Moisture Sensor Testing
In this research, the author will test the capacitive soil moisture sensor to determine the error value, accuracy, and precision of the sensor. The test was carried out by comparing the humidity values released by the comparison device (moisture meter) and the capacitive soil moisture sensor. In measuring instruments, the range of humidity values is 1 to 10, but in testing the value 1 to 10 is assumed to be 10% to 100%. Tests were carried out in three different conditions, namely when the soil moisture level was wet (90%), neutral (40%), and dry (20%), where 30 data samples were taken for each condition. The results of soil moisture testing can be seen in Table 4
Table 4. Soil Moisture Testing Measurement Soil
Moisture (%)
AVG Moisture meter (%)
Avg Soil moisture Sensor (%)
Avg Difference
(%)
Avg Error
(%)
Avg Accuracy
(%)
Avg Precision
(%)
20 20 21,667 1,67 8,33 91,67 97
40 44 41,83 1,83 4,58 95,42 99
90 90 88,67 1,3 1,48 98,52 99
Based on Table 4, it is known that the results of the 3 soil moisture samples have measurement results with an accuracy of above 90% and an error below 10%. Based on measurement standards, it can be stated that the results are not good enough because they have a measurement error of more than 5 percent, although they can still be tolerated. After all, the precision results are above 97% with 30 testing samples.
3.5 End-to-End Testing
Overall tool testing aims to find out whether the tool can work as it should based on the system created. In this system, the tool is designed to detect soil temperature values, soil pH values, and soil moisture. Apart from that, the system created also has a water pump that works as an automatic waterer for the NPK fertilizer solution when the condition or pH value of the soil is less than 5.5.
Testing of the entire system was carried out directly on the cucumber plantation in Binangun District, Cilacap Regency. During implementation, all sensors used are placed in the ground, while the water pump is connected to two hoses. The first hose is connected to a container containing the fertilizer solution to drain the solution, while the second hose is placed directly into the soil to distribute the solution to the plants. In the automatic fertilizer solution watering system that was created, the watering can not yet flow the solution throughout the land, it can only flow the solution per planting hole. When the tool is ready and the power source is connected to the tool, the LCD will immediately display "Starting system and Network" which means the tool starts the system and checks the network to send data to the IoT Platform. After the process is complete, the LCD will display the soil temperature value, soil pH value, and soil moisture value, and display the status or condition of the pump. The pump condition status is represented by the number 0 or 1. When the pump condition is 1 then the pump is on, while when the pump condition is 0 then the pump is not on. Next, after taking data for one minute, the last data detected by the system will be sent to the Telkom IoT Platform. The LCD will display "Sending data to Platform". Figure 7 shows the overall system test on a cucumber plantation and the test data results can be seen in table 5.
Figure 8. Whole System Testing
18/06/2023 13:00 6,73 29,69 65 OFF
18/06/2023 13:01 6,76 29,75 64 OFF
18/06/2023 12:52 6,8 29,88 65 OFF
18/06/2023 13:02 5,41 29,81 65 ON
18/06/2023 13:03 5,31 29,81 64 ON
18/06/2023 13:04 5,05 29,81 65 ON
18/06/2023 13:05 5,13 29,81 64 ON
18/06/2023 13:06 5,34 29,88 64 ON
18/06/2023 13:07 5,28 29,88 64 ON
18/06/2023 13:08 5,06 29,81 65 ON
18/06/2023 13:09 5,25 29,88 65 ON
18/06/2023 13:10 5,17 29,88 65 ON
18/06/2023 13:11 5,3 29,81 65 ON
Based on the data listed in Table 5, it can be concluded that the overall system testing has been successful.
The results of this test reveal that the system can operate effectively and provide satisfactory performance.
Measurement data sent to the IoT Platform can be seen in figure 9. Figure 9 shows that data read by the system can be entered and stored on the Telkom IoT Platform. The data at the top is the data that was last entered into the IoT Platform. Data read by the system is sent once a minute to the IoT platform.
Figure 9. Data On IoT Platform
4. CONCLUSION
Based on the results of design and testing, the control and monitoring system for planting media for cucumber plants can work well according to the system created because the data can be stored on an IoT Platform. Apart from that, the average temperature sensor error value was 0.68%, the average accuracy was 99.32%, and the average precision was 0.99. Meanwhile, the results of testing the soil moisture sensor had an average error of 4.77%, an average accuracy of 95.32%, and a precision of 0.98. Based on the test results, it can be said that the temperature sensor and humidity sensor can read soil temperature and humidity values well. Then the pH sensor
test results using the linear regression method, obtained an average error value of 1.65%, an average accuracy value of 98.35%, and an average precision value of 0.98. So it can be said that the sensor can read the soil pH value well.
ACKNOWLEDGMENT
This research is supported by several parties, including the Telkom Purwokerto Institute of Technology and Telkom Corporate University
REFERENCES
[1] S. Idris, N. Musa, and W. Pembengo, “Produksi Tanaman Mentimun (Cucumis sativus L.) Akibat Pemangkasan Dan Jumlah Benih Per Lubang Tanam,” Jatt, vol. 7, no. 2, pp. 229-235 ISSN 2252-3774, 2018, [Online]. Available:
https://repository.ung.ac.id/skripsi/show/613411010/produksi-tanaman-mentimun-cucumis-sativus-l-akibat- pemangkasan-dan-jumlah-benih-per-lubang-tanam.html
[2] A. Sihaloho, R. Purba, and D. E. Siregar, “PENGARUH PUPUK BIOORGANIK DAN DOSIS PUPUK NPK MUTIARATERHADAP PERTUMBUHAN DAN PRODUKSI TANAMAN MENTIMUN (Cucumis sativus L.),” J.
Rhizobia, vol. 8, no. 1, pp. 32–41, 2020, doi: 10.36985/rhizobia.v8i1.70.
[3] R. Zikria and A. Damayanti, “Peran Penyuluhan Pertanian dan Preferensi Risiko terhadap Penggunaan Pupuk Berlebih pada Usaha Tani Padi,” J. Agro Ekon., vol. 37, no. 1, p. 79, 2019, doi: 10.21082/jae.v37n1.2019.79-94.
[4] R. G. Wisduanto, A. Bhawiyuga, and D. P. Kartikasari, “Implementasi Sistem Akuisisi Data Sensor Pertanian Menggunakan Protokol Komunikasi Lora,” J. Pengemb. Teknol. Inf. dan Ilmu Komput., vol. 3, no. 3, pp. 2201–2207, 2019.
[5] A. Sharma, P. K. Singh, and Y. Kumar, “An integrated fire detection system using IoT and image processing technique for smart cities,” Sustain. Cities Soc., vol. 61, p. 102332, 2020, doi: 10.1016/j.scs.2020.102332.
[6] S. Pamungkas, “Smart Greenhouse System On Paprican Plants Based On Internet of Things,” Telekontran J. Ilm.
Telekomun. Kendali dan Elektron. Terap., vol. 7, no. 2, pp. 197–207, 2020, doi: 10.34010/telekontran.v7i2.2277.
[7] A. Nazarudin, M. Mahdiannoor, and Z. Zarmiyeni, “Pertumbuhan dan Produksi Tanaman Mentimun terhadap Pemberian Berbagai Takaran Vermikompos pada Tanah Podsolik Merah Kuning,” Rawa Sains J. Sains Stiper Amuntai, vol. 9, no.
1, pp. 705–714, 2019, doi: 10.36589/rs.v9i1.95.
[8] A. F. Zulkarnain, E. S. Wijaya, and N. F. Mustamin, “Penerapan Teknologi Smart Farming Berbasis Internet of Things Bagi Masyarakat Petani Jeruk Siam,” Batara Wisnu Indones. J. Community Serv., vol. 2, no. 1, pp. 50–59, 2022, doi:
10.53363/bw.v2i1.47.
[9] A. Tzounis, N. Katsoulas, T. Bartzanas, and C. Kittas, “Internet of Things in agriculture, recent advances and future challenges,” Biosyst. Eng., vol. 164, no. December, pp. 31–48, 2017, doi: 10.1016/j.biosystemseng.2017.09.007.
[10] Q. Wen and Q. Chen, “The application of Internet of things technology in pharmaceutical cold chain logistics,” WIT Trans. Inf. Commun. Technol., vol. 61, no. 7, pp. 617–624, 2014, doi: 10.2495/MIIT130791.
[11] L. Jiang, L. Da Xu, H. Cai, Z. Jiang, F. Bu, and B. Xu, “An IoT-Oriented data storage framework in cloud computing platform,” IEEE Trans. Ind. Informatics, vol. 10, no. 2, pp. 1443–1451, 2014, doi: 10.1109/TII.2014.2306384.
[12] H. S. Nida, M. Faiqurahman, and Z. Sari, “Prototype Sistem Multi-Telemetri Wireless Untuk Mengukur Suhu Udara Berbasis Mikrokontroler ESP8266 Pada Greenhouse,” Kinet. Game Technol. Inf. Syst. Comput. Network, Comput.
Electron. Control, vol. 2, no. 3, pp. 217–226, 2017, doi: 10.22219/kinetik.v2i3.89.
[13] I. Kuzminykh, A. Snihurov, and A. Carlsson, “Testing of communication range in ZigBee technology,” 2017 14th Int.
Conf. Exp. Des. Appl. CAD Syst. Microelectron. CADSM 2017 - Proc., pp. 133–136, 2017, doi:
10.1109/CADSM.2017.7916102.
[14] A. Arora and A. Grover, “ZigBee: Simulation and Investigation of Star and Mesh Topology by Varying Channel Sensing Duration,” Int. J. Comput. Appl., vol. 67, no. 9, pp. 44–50, 2013, doi: 10.5120/11425-6771.
[15] M. L. Liya and M. Aswathy, “LoRa technology for Internet of Things(IoT): A brief Survey,” Proc. 4th Int. Conf. IoT Soc. Mobile, Anal. Cloud, ISMAC 2020, pp. 128–133, 2020, doi: 10.1109/I-SMAC49090.2020.9243449.
[16] A. Bhardwaj, K. Kaushik, S. Bharany, M. F. Elnaggar, M. I. Mossad, and S. Kamel, “Comparison of IoT Communication Protocols Using Anomaly Detection with Security Assessments of Smart Devices,” Processes, vol. 10, no. 10, 2022, doi:
10.3390/pr10101952.
[17] V. Sarafov and J. Seeger, “Comparison of IoT Data Protocol Overhead,” Semin. Futur. Internet SS2017, Dep.
Informatics, Tech. Univ. Munich, no. March, pp. 7–14, 2018, doi: 10.2313/NET-2018-03-1.
[18] A. Wardhana et al., Arsitektur dan standarisasi internet of things (iot), no. May. 2023.
[19] G. Santoso, S. Hani, and U. D. Putra, “Monitoring kualitas tanah lahan pertanian Desa Sidorejo menggunakan sensor pH tanah dan Internet of Things,” J. Nusant. Mengabdi, vol. 2, no. 1, pp. 1–10, 2022, doi: 10.35912/jnm.v2i1.1387.
[20] R. Daniel, “Rancang Bangun Alat Monitoring Kelembaban, PH Tanah dan Pompa Otomatis Berbasis Arduino,” J. Appl.
Comput. Sci. Technol., vol. 3, no. 2, pp. 208–212, 2022, doi: 10.52158/jacost.v3i2.384.
