Predicting Sodium Concentration in River Water Using Novel Hybrid Deep Learning Models

This study developed a hybrid model consisting of two deep learning algorithms for predicting concentrations of minerals such as sodium in rivers. The hybrid model incorporated convolutional neural networks (CNNs) and long short-term memory networks (LSTMs). The model was trained using a robust dataset from 2013 to 2023 for the water quality in Georgia streams. The results showed a successful training process for the hybrid CNN–LSTM model. The CNN–LSTM model indicated a high accuracy in the prediction of minerals in rivers compared to the regular machine learning models. With a coefficient of determination of greater than 0.9 in all simulations and the relative error less than 10%, the CNN–LSTM model outperformed the regular models. By an integrated data preprocessing approach and reducing the need for extensive manual intervention, the hybrid CNN–LSTM model was shown to be a highly accurate predictive tool for monitoring water quality for rivers.