Predictive modeling of ash and moisture content in acid-treated walnut shells using artificial neural networks

In this paper, the use of artificial neural networks (ANN) to predict changes in ash and moisture content (MC) of walnut shells after acidic pretreatment was investigated. The physicochemical and structural properties of walnut shells were analyzed, as well as the influential parameters on the changes of these studied variables. The study aimed to determine the changes in the walnut shells composition and evaluate the possibility of using the ANN model to predict the above-mentioned initial variables. The ANN model showed extremely high performance in modeling, with a coefficient of determination (R²) of 0.99 for ash and 0.90 for MC and low values of root mean square error (RMSE) and χ², indicating the accuracy of the model in predicting the composition of walnut shells after acidic pretreatment. This study confirms the potential of ANN in accurately modeling changes in the composition of walnut shells and paves the way for further research on a wider range of crops and different pretreatments.