(JAIS) Journal of Applied Intelligent System

Application of Deep Learning in Bone Digital Densitometer Based on Electrical Impedance Spectroscopy

Sari Ayu Wulandari — Thu, 22 Oct 2026 00:00:00 +0000

- In Indonesia, over 23% of women between the ages of 50 and 80 suffer from osteoporosis, a serious bone health condition. This condition may lead to fractures that affect quality of life. The disadvantages of traditional bone mineral density (BMD) measurement methods, like DEXA and QCT, include high costs and radiation exposure. By combining Deep Learning technology with Electrical Impedance Spectroscopy (EIS), this study seeks to provide a different, safer, and more portable approach. A total of 12,480 measurement data were obtained by collecting EIS data from four levels of bone density (30%, 50%, 70%, and 90%).This process involves signal pre-processing using end point detection techniques and artificial feature extraction. Five machine learning algorithms Decision Tree, Naive Bayes, Support Vector Machine, Logistic Regression, and XGBoost are used for bone density classification. As a result, the integration of EIS with machine learning algorithms has been shown to improve the accuracy of osteoporosis detection, where XGBoost shows the best performance and is able to overcome overfitting. This innovation is expected to expand access to osteoporosis diagnosis in health facilities, especially in developing countries like Indonesia.

Automatic Drip Irrigation System Based on ESP32 Microcontroller Using Average Soil Moisture Soil Moisture Value for Melon Plants

Nor Amir Mahmud, Noor Yulita Dwi Setyaningsih — Thu, 22 Oct 2026 00:00:00 +0000

Maintaining optimal soil moisture is crucial for supporting melon cultivation in greenhouses and ensuring efficient irrigation management. However, conventional drip irrigation is typically carried out manually or based on readings from a single soil moisture sensor, which may not accurately reflect soil moisture conditions across the entire cultivation area. This study aims to develop an automated drip irrigation system based on Average Soil Moisture for melon cultivation in greenhouses. The proposed system consists of one ESP32 controller and six ESP32-based sensor nodes equipped with capacitive soil moisture sensors that communicate via the ESP-NOW protocol. The controller continuously calculates the Average Soil Moisture and applies a hysteresis control algorithm by activating the irrigation pump when the Average Soil Moisture drops below 75% and deactivating it when it reaches 80%. The developed system also provides real-time local monitoring via an OLED display. The developed soil moisture sensing system achieves a measurement error ranging from 1.3% to 4.6%, with measurement accuracy between 95.4% and 98.7%. This irrigation system is capable of maintaining the Average Soil Moisture within the target operating range of 75%–80%, with an average error of 0.21% relative to the lower setpoint, a maximum overshoot of 0.6%, and an average error of 0.41% relative to the upper setpoint, demonstrating the stability of hysteresis-based irrigation control. Overall, the proposed control strategy based on Average Soil Moisture provides more representative irrigation decisions compared to single-point measurements, thereby resulting in stable automatic drip irrigation performance and improving irrigation reliability for melon cultivation in greenhouses.

Design and Performance Analysis of an Automatic Solar PV–Grid Power Transfer System Based on Battery Voltage for the Hidroguard System

Ferry Aldi Ardiansyah, Noor Yulita Dwi Setyaningsih — Thu, 22 Oct 2026 00:00:00 +0000

The Hidroguard system requires a stable supply of electrical energy to operate the circulation pump, sensors, microcontrollers, and monitoring devices continuously. The use of the Solar Power Plant (PLTS) system as the main source has limitations because energy production is affected by weather conditions and the energy stored in the battery. This study analyzes Low Voltage Disconnect (LVD) as a controller of automatic energy transfer from the PLTS-battery line to the PLN network through a 12 VDC adapter. The system uses a 240 Wp monocrystalline solar panel and a 12 V 150 Ah LiFePO4battery. The experimental method was carried out by observing the battery voltage, LVD contact conditions, relay work, and active energy sources. The LVD set point is set at 11 V. The PV line remains active at a battery voltage of 13.2 V to 11 V. When the voltage is recorded to drop to 10.8 V, the LVD disconnects the battery line and shifts the Hydroguard load to the PLN network through an adapter. The load returned to the solar power line after the battery voltage increased to 11.5 V. During the 48-hour observation, the application of LVD was able to maintain the continuity of energy supply while protecting the battery from over-discharging conditions. Based on the six conditions of energy source transfer testing, the entire transfer process takes place according to the control logic so that a system success rate of 100% is obtained.

IOT Web-Based Multinode Weather Monitoring System Using Iot

Bagus — Thu, 22 Oct 2026 00:00:00 +0000

This research presents a real-time web-based multinode weather monitoring system using an Internet of Things (IoT) architecture. The system is designed to address the need for continuous and accurate environmental observation across multiple locations, particularly for supporting decision-making in agriculture, disaster mitigation, and smart-city applications. Three sensor nodes equipped with temperature, humidity, air pressure, light intensity, and rainfall sensors were developed using the ESP8266 microcontroller. Each node transmits environmental data to the Firebase Realtime Database at five-minute intervals, supported by an RTC module for precise timestamping and a micro-SD card for backup storage. A web-based dashboard was developed to display real-time data in the form of graphs, tables, and condition indicators for each monitoring site. The results show that the multinode system operates reliably, providing consistent sensor readings, stable data transmission, and synchronized visualization on the web dashboard. Although the rainfall sensor demonstrated occasional inaccuracies due to surface contamination, overall system performance remained stable throughout testing. This study concludes that the proposed IoT-based multinode system offers an effective, scalable, and low-cost solution for real-time weather monitoring. Future improvements may include enhanced sensor accuracy, additional environmental parameters, and predictive analytics for advanced environmental assessment.

Keywords - Internet of Things, weather monitoring, sensor nodes, real-time system, web-based dashboard

Web-Based Real-Time Monitoring System for TDS, Temperature, Humidity, and Cocopeat Moisture in Greenhouse Hydroponic Melon Cultivation

Elisa, Noor Yulita Dwi Setyaningsih — Thu, 22 Oct 2026 00:00:00 +0000

Environmental monitoring is crucial for maintaining suitable greenhouse conditions for melon cultivation. Conventional monitoring methods require manual observation and are unable to provide continuous real-time information. This study developed a web-based greenhouse monitoring system using ESP32 and ESP32-C3 Supermini microcontrollers with ESP-NOW wireless communication. The system integrates capacitive soil moisture sensors, Total Dissolved Solids (TDS), and reed switch sensors distributed across six ESP32-C3 sensor nodes. Sensor data is transmitted to an ESP32 gateway via ESP-NOW and then uploaded to a hosted web server via a Wi-Fi connection. The web application was developed using PHP, HTML, CSS, JavaScript, and MySQL, providing a real-time dashboard, historical data monitoring, graphical visualization, pagination, and Microsoft Excel export features. Communication testing demonstrated a transmission delay of 0–1 second under good network conditions without data loss, while the delay increased to 2–3 seconds with occasional transient packet loss under poor network conditions. Overall, the proposed system successfully enables reliable real-time and remote monitoring of greenhouses, providing an effective solution to support greenhouse management and precision agriculture.

Microcontroller-Based Automatic pH Control System with IoT-Based Remote Monitoring on Melon Plant NFT Hydroponics

Muhammad Fadilan Maulana, Noor Yulita Dwi Setyaningsih — Thu, 22 Oct 2026 00:00:00 +0000

Abstract - Hydroponic cultivation using the Nutrient Film Technique (NFT) circulates a thin, continuous film of nutrient solution through the growing channel, making the stability of the solution's potential of hydrogen (pH) a critical factor in determining nutrient solubility and root uptake efficiency. Conventional pH monitoring that relies on manual measurement using a pH meter is inefficient, labor-intensive, and prone to delayed handling due to human error, so pH fluctuations occurring outside the inspection schedule cannot be detected promptly. This study aims to design and evaluate an automatic pH monitoring and control system for the circulation of an NFT-type hydroponic system based on the Internet of Things (IoT). The system uses a pH sensor on a Soil 7-in-1 module connected to an ESP32 microcontroller as the main controller, which processes the sensor reading to activate pH-up and pH-down dosing pumps through a relay module based on predefined threshold values. The reading and control status are displayed on an LCD and a web-based dashboard for real-time, remote monitoring. Testing was carried out in two stages: evaluating the accuracy of the pH sensor against a calibrated pH meter, and evaluating the performance of the pH control system during continuous solution circulation. The results show that the pH sensor achieves an accuracy of 99.2% with an average error of 0.8%, while the control system is able to maintain the pH value within the setpoint range of 5.8-7.2 with an average response time of 30-60minutes, overshoot of 2.8% and an undershoot of 6.9%. Data transmission delay testing was also conducted under two signal conditions, namely good signal and poor signal. Good signal conditions, the system achieved an average transmission delay of 0.27 seconds, while poor signal conditions, the average delay increased to 1.08 seconds due to reduced network quality. Nevertheless, all data were successfully transmitted under both conditions, indicating that the system remains reliable even when signal quality deteriorates. The developed system is shown to be capable of automatically and continuously maintaining the stability of the nutrient solution's pH, offering a practical solution to reduce manual intervention in NFT hydroponic cultivation.