Intelligent Monitoring System for Fry Rearing Using Fuzzy Logic

This study addresses the development of an intelligent monitoring system for water quality in rainbow trout (Oncorhynchus mykiss) breeding pools. The system is based on IoT (Internet of Things) technology and Artificial Intelligence, specifically employing the fuzzy logic algorithm. Three submersible electronic sensors are utilized to capture readings and interpret environmental conditions and water quality parameters. These sensors were selected based on their relevance to the species’ production, including pH concentration levels, sudden temperature changes, and oxygen saturation levels in the water. The collected data is gathered by two sensor nodes and transmitted wirelessly to a coordinator node through an IEEE 802.15.4 protocol-based wireless sensor network (WSN). A web platform is employed to visualize the collected data, enabling the manager to monitor the water quality in the ponds and receive alerts in case of adverse events. To facilitate event notifications, both on the web platform and via Telegram, a fuzzy logic algorithm is implemented. This algorithm incorporates 27 fuzzy rules, derived from the expertise of laboratory specialists. The algorithm demonstrates an accuracy of 94.4% and a precision of 100% as determined by the confusion matrix.