Wireless Sensor Sub-Network Based IoT System for Probiotic Dosing and Water Quality Management Using Artificial Neural Network

Water quality management in aquaculture plays a crucial role in maintaining fish health and optimizing growth, particularly in intensive tilapia farming. This study develops a Wireless Sensor Sub-Network (WSSN) based Internet of Things (IoT) system designed to automate probiotic dosing and monitor water quality conditions using real-time sensor feedback and Artificial Neural Network (ANN) analysis. Utilizing TCS3200 color sensors, flow sensors, and an ANN within a WSSN, the system autonomously manages probiotic delivery based on real-time water color analysis, marking a shift towards intelligent water-quality management in aquaculture. The system architecture consists of three primary sensing and control components: a flow sensor connected to an ESP32 microcontroller to measure the precise volume (in milliliters) of probiotic solution dispensed; a TCS3200 color sensor, also integrated with an ESP32 module, to detect variations in water color as an indicator of pond health; and a solenoid valve controlled through a relay-actuated ESP32 node to regulate probiotic release into the pond. The sensor network operates wirelessly to provide continuous monitoring and intelligent decision-making. The ANN employs the backpropagation algorithm to perform color-based classification, where light green indicates a healthy condition, dark green represents normal stability, light brown signals the need for probiotic dosing, and dark brown denotes a critical condition requiring water replacement. This integration of optical and flow sensing with neural network computation provides an intelligent, non-invasive, and adaptive mechanism for probiotic management in tilapia aquaculture, supporting sustainable aquaculture practices and improving operational efficiency through automation and predictive learning.