REAL-TIME AQUACULTURE MONITORING SYSTEM USING AIOT TECHNOLOGY

Abstract

Aquaculture contributes greatly to the overall fish production, but old methods of monitoring water quality prove ineffective and continue to allow high fish mortality and low productivity. This paper presents a Real-Time Artificial Internet of Things (AIoT) Monitoring System for continuous monitoring of critical water parameters in aquaculture that would enhance management parameters such as temperature, turbidity, dissolved oxygen (O₂), pH, CO₂ and water levels. The system represents the integration of many sensors with an ESP8266 microcontroller which transmits data to a cloud-based platform that can be accessed by a mobile application. Machine learning techniques such as Decision Trees, Support Vector Machines, and Logistic Regression will be used to optimize the activation of the used actuators (oxygen generators, feeders, water pumps, and heaters) based on real-time detected measurements from sensors. Experimental results prove that it works; steady monitoring and precise prediction of best water conditions correspond to what is theoretically expected. The AI-powered decision-making framework that curtails energy usage and feed wastage plus guarantees a prompt reaction to changes in the environment renders this a smart aquaculture solution: scalable and cost-effective for modern fish farming while fostering enhanced productivity and sustainability for the aquaculture sector.