Application of Artificial Neural Network in Predicting the Drying Kinetics and Chemical Attributes of Linden (Tilia Platyphyllos Scop.) During the Infra-Red Drying Process

Abstract

This study investigates the potential of applying artificial neural networks (ANNs) to describe the drying kinetics of linden leave samples during infra-red drying (IRD) under different drying temperatures of 50&DEG;C, 60&DEG;C and 70&DEG;C and samples thickness (0.210, 0.220, and 0.230). Kinetic models were developed using selected thin layer models, and ANN methods. The statistical indicators of the coefficient of determination (R2), and root mean square error (RMSE) were used to evaluate the suitability of the models. The effective moisture diffusivity varied between 4.13 x 10-12 m2/s and 5.89 x 10-12 m2/s and the activation energy was 16.339 kJ/mol. The thin-layer models illustrated that all used models (Page, Midilli et al., Henderson and Pabis, Logarithmic, and Newton models) can adequately describe the drying kinetics of linden leave samples with R2 values (> 0.9900) and lowest RMSE (<0.0200). The ANN model showed R2 and RMSE values of 0.9986, and 0.0210, respectively. Also, the ANN model shows the significant prediction for the linden chemical attributes for Total phenolics content (TPC), Total flavonoids assay (TFA), DPPH, and FRAP of R2 and RMSE values of 0.9975, 2.6100, 0.9891, 0.1346, 0.9980, 2.9317, 0.9845, and 0.9808, respectively. © 2022 TAE 2022 - Proceeding of the 8th International Conference on Trends in Agricultural Engineering 2022. All rights reserved.