AquaFeed: um sistema de monitoramento dos parâmetros Físico-Químicos e recomendação para auxílio do manejo alimentar na Piscicultura

Abstract

Fish farming is a source of income for many farmers, who still lack technologies to help with the daily management of ponds, especially with regard to monitoring water quality. This work proposes the development of a system to monitor the physical and chemical parameters of water (temperature, pH, dissolved oxygen and electrical conductivity) and to help with food management in tambaqui fish farming operations. The system uses machine learning algorithms, applying regression techniques to generate daily feeding recommendations. The system was developed with hardware to collect and monitor the parameters and software to visualize the data and generate recommendations. The system was evaluated in two stages. Laboratory tests evaluated the behavior of the sensors and the accuracy of the measurements. The results indicated that the sensors keep up with sudden changes. Field tests compared the system's performance with commercial sensors. To obtain the results, the F-tab test was applied. The results showed that the system presented results that were statistically close to the commercial sensors, with calculated F values, FcpH = 1.87, FcTemperature = 1.68, FcElectrical conductivity = 7.84.