• Mr.P.Yejdani Khan
• K.Hemanth Kumar

Predicting water quality (WQP) is crucial for effective management in aquaculture, ensuring both profitability and sustainability of production. This study introduces hybrid deep learning (DL) models, combining convolutional neural networks (CNN) with long short-term memory (LSTM) and gated recurrent units (GRU) for WQP in aquaculture. The CNN efficiently captures the characteristics of aquaculture water quality, while the GRU and LSTM models excel at learning long-term dependencies in time series data. We conducted experiments on two different water quality datasets and performed a comprehensive analysis of how hyperparameters affect the performance of the proposed hybrid DL models. Additionally, we compared the performance of the hybrid CNN-LSTM and CNN-GRU models against various baseline LSTM, GRU, and CNN DL models, as well as attention-based LSTM and attention-based GRU DL models. The findings indicate that the hybrid CNN-LSTM model surpassed all others in prediction accuracy and computation time.

Aquaculture, Convolutional Neural Networks (CNN), deep learning, Gated Recurrent Units (GRU), hybrid models, Long Short-Term Memory (LSTM), water quality forecasting.

"Water Quality Prediction For Smart Aquaculture Utilizing Hybrid Deep Learning Techniques", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 6, page no.319-327, June-2024, Available :http://www.jetir.org/papers/JETIRGL06054.pdf

"Water Quality Prediction For Smart Aquaculture Utilizing Hybrid Deep Learning Techniques", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 6, page no. pp319-327, June-2024, Available at : http://www.jetir.org/papers/JETIRGL06054.pdf