Development of a Water Tank Monitoring and Pump Control System Using ESP32 and the ThingsBoard Platform with Telegram Notifications at PDAM Kota Surabaya

Abstract

Conventional water tank monitoring at PDAM Surya Sembada Kota Surabaya causes delayed information and risks of overflow or underflow. This study develops an IoT-based monitoring and automatic pump control system utilizing an ESP32 microcontroller, an HC-SR04 ultrasonic sensor, the ThingsBoard platform via MQTT, and Telegram notifications. A single ultrasonic sensor simultaneously derives water level, cylindrical volume, and flow rate (Q = ΔV/Δt), eliminating physical flowmeters. Pump control is automated using two-threshold hysteresis logic. Experimental results demonstrate a water level measurement accuracy of 98.63% (MAE = 0.14 cm, R² = 0.9998). The derived flow rate was empirically validated against manual measurements, achieving 99.26% accuracy with an error of only 0.06 L/min. Telemetry latency averaged 0.53 seconds on ThingsBoard, while Telegram early warning alerts were delivered within 2.38 seconds. Furthermore, a three-layer security architecture successfully thwarted unauthorized access scenarios, and expert validation confirmed a 92.0% feasibility rate (Highly Feasible). Thus, the proposed system provides an accurate, robust, and cost-effective solution for municipal water distribution facilities.