Real-Time Boiler Feed Water Quality Assessment System Based on Internet of Things and Fuzzy Sugeno Logic

Abstract

The quality of boiler feed water is a critical factor influencing the performance and reliability of marine steam boilers. Poor water quality can cause corrosion, scaling, and reduced operational efficiency. This study designed and implemented a real-time boiler feed water quality assessment system using NodeMCU ESP32 integrated with Internet of Things (IoT) connectivity and fuzzy Sugeno logic for decision-making. The system collected data from turbidity, salinity, and pH sensors, transmitted them to Firebase, and displayed the results on a 0.96-inch Organic Light-Emitting Diode (OLED) screen. A zero-order Sugeno fuzzy inference system classified water quality into good, moderate, and poor categories. Testing on seven water samples achieved an accuracy of 92.97%. Compared to conventional monitoring systems, the integration of IoT and fuzzy inference provides a novel, practical, and computationally efficient approach for real-time water quality monitoring in maritime boiler applications.