Water Quality Classification Using Multi-cell RNN in Aquaculture Ponds for Catla Fish

In aquaculture, water quality has a crucial role in fish development and survival. The link between different water quality sections is not well captured by the water quality prediction systems now in use, and their capacity to take advantage of the temporal relationships and interactions between various physical and chemical properties of water parameters is constrained. In this study, we present a multi-cell RNN for catla fish that combines the benefits of LSTM and GRU to capture these temporal linkages and interactions in a deep learning model for water quality prediction in aquaculture catla fishponds. Compared to other models, our suggested approach performs better at categorizing data on water quality for catla fish. The experimental findings show that the suggested multi-cell RNN achieves a high accuracy of 99.88% and a low loss of 0.0192. This study offers a viable method for forecasting aquaculture water quality, which may help fish develop and survive and support sustainable aquaculture methods.