Answers are grounded in this paper and cite their sources.
Rare Earth Elements (REEs) are of great importance in various high-tech industries, including green tech, advanced electronics, and defense systems. In this study, a set of Artificial Neural Network (ANN) models, based on artificial intelligence techniques, was developed and experimentally validated to predict the remaining REE in phosphogypsum ( P m r e e o u t ) , as well as the REE passage rate in the leachate ( P R E E r l ) , processed at Jorf Lasfar, Morocco. The models use measured values of phosphogypsum mass ( P m g y d r ) , REE concentration ( P r e e i ) , and REE mass ( P m r e e i ) as input parameters. Two nonlinear activation functions were applied in the hidden layer (Logsig and Tansig), while a linear function (Purelin) was used in the output layer. The Levenberg–Marquardt algorithm was employed to train the ANN models. The results demonstrated that the developed ANN models can accurately estimate P m r e e o u t , with a coefficient of determination of 0.9983 and a Root Mean Square Error (RMSE) of 0.0006, as well as ( P R E E r l ) , with a coefficient of determination of 0.9548 and an RMSE of 3.0026, highlighting their potential for the future optimization of extraction processes.
No takes yet. Share an insight, caveat, or question.
Ajbar et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: