Abstract:
This research aims to develop a real-time flood early warning system using IoT technology and a LoRaWAN network for deployment in rural areas frequently affected by recurrent floods. The developed prototype system consists of three ultrasonic water level sensors and a rain gauge sensor, connected to a microcontroller that transmits data via the LoRaWAN network to a Raspberry Pi gateway. The gateway records the data onto a cloud database (Google Sheets) and displays the results on a real-time dashboard via Looker Studio. The system is powered by a solar panel coupled with a battery, enabling continuous operation even during power outages. It automatically sends alerts through the LINE Application when the water level exceeds a predefined threshold, warning schools and communities to prepare for the flood hazard in advance. Experimental results, gathered over a random 20-day period, demonstrate that the system can continuously and accurately measure water levels and rainfall amounts, achieving an average error of 7%. The standard deviation of the data collected over the 20 days was 39.71. Data was transmitted in real-time, and the flood alerts were delivered promptly, resulting in schools and communities receiving information faster to mitigate flood damage. The developed system is low-cost, low-power consumption, and offers reliable long-range communication. Furthermore, it is easy to maintain and excellently supports deployment in remote areas with limited communication infrastructure. It significantly helps reduce the impact of floods and enhances the efficiency of water resource management and disaster response in the study area.