Telegram bot-based Flood Early Warning System with WSN Integration

Abstract

Indonesia experiences frequent flooding, with data from the National Disaster Management Agency (BNPB) revealing that floods account for 41% of all natural disasters (1,441 incidents) recorded in 2021. These floods cause significant property damage and casualties. To address this challenge, we have developed a prototype flood early warning system. This system utilizes ultrasonic sensors for real-time water level detection. Sensor data is transmitted to designated personnel through a website interface. Additionally, the system leverages a Telegram bot to deliver flood early warnings directly to the community residing in flood-prone areas. The sensor data comparison test yielded an error rate of only 0.6175% with an average difference of 1 cm, demonstrating the system's accuracy and functionality. Furthermore, a notification test conducted ten times achieved 100% accuracy. The Telegram bot successfully sent text message alerts (alert 1, alert 2, alert 3) with an average delivery time of 4.07 seconds. This prototype offers a promising solution for flood mitigation. By providing real-time water level data and issuing timely alerts via a user-friendly Telegram bot, the system empowers communities to prepare for potential flooding and minimize associated risks.

Keywords

• alert
• early warning system
• flood
• water level detection
• wireless sensor network