| dc.description.abstract | Forest fires pose a serious threat to ecosystem sustainability and human safety, particularly in Indonesia, which possesses extensive tropical forest areas. This study aims to design and evaluate an early forest fire detection system based on the Internet of Things (IoT) using a combination of the SHT31-D temperature sensor and the UVTron R2868 flame sensor. The system is equipped with a Neo 6M GPS module for determining fire location coordinates, a NodeMCU ESP32 microcontroller for data processing, and the ThingSpeak platform alongwith Telegram for real-time data storage and notification delivery. The system is powered by three 18650 batteries, while a buzzer alarm provides local alerts. The research methodology includes hardware design, programming, sensor calibration, and field testing under three environmental conditions: forest interior, residential area, and roadside. This detection system is programmed with a temperature threshold of 40°C as an early warning indicator of potential forest fires. The SHT31-D sensor demonstrates an accuracy of ±0.39°C with an average percentage error of approximately 2%. Average temperature measuerements were 35,68°C in the forest interior, 35,175°C in the residential area, and 34,37°C at the roadside. During early fire conditions, the temperature reached 37,59 °C in the forest interior, 36,4°C in residential areas, and 35°C near the roadside. Experimental results of the UVTron R2868 sensor show its ability to detect flame sources from a lighter at a distance of up to 9 meters. Test results demonstrated that the UVTron R2868 sensor could detect flames at distances
of 7 meters in the forest interior and residential areas, and 8 meters at the roadside. The system succesfully transmits temperature data, fire detection status, and GPS coordinates rapidly and accurately within only 1,13 minutes, while minimizing false alarms through a dual-sensor mechanism. This research demonstrates that the integration of IoT technology can enhance the effectiveness of early forest fire detection, thus supporting more timely mitigation and response efforts. | en_US |
