Design and Development of an Intelligent IoT-Based Monitoring System for Biofloc Fish Farming Using Machine Learning

Aquaculture requires continuous monitoring of water quality parameters to ensure healthy fish growth and improved productivity. Traditional monitoring methods are manual, time-consuming, and unable to detect sudden environmental changes. This paper presents the design and development of an intelligent IoT-based monitoring system integrated with machine learning for biofloc fish farming. The system measures key parameters such as temperature, humidity, pH, turbidity, and total dissolved solids (TDS) using multiple sensors interfaced with an Arduino Mega microcontroller. The collected data is transmitted via the ESP8266 Wi-Fi module to an IoT cloud platform for real-time monitoring and visualization. A Random Forest algorithm is employed to analyze the data and predict abnormal water conditions. Based on the analysis, an automated relay mechanism activates a water pump to maintain optimal water quality. The proposed system improves efficiency, reduces manual effort, enables predictive decision-making, and enhances overall productivity in aquaculture systems.