Mathematical and artificial neural network modelling for refractance window drying kinetics of coriander (Coriandrum sativum L.) followed by the determination of energy consumption, mass transfer parameters and quality

Coriander (Coriandrum sativum L.) is an aromatic and nutritious herb but often wasted due to limited processing. Drying is a common method for extending its shelf life. In the study, coriander was hot water blanched, pureed and dried with varying thicknesses of puree (2, 4 and 6 mm) and water temperatures (70, 80 and 90 °C) to assess drying characteristics, mass transfer and quality of coriander powder. It was noted that the time required for drying decreased as the water temperature was raised from 70 to 90 °C. The mathematical modelling showed that the Exponential two-term model had the highest R2 and lowest RSME and SEE values. Moreover, MR was accurately predicted using MLF-ANN with back-propagation algorithm, outperforming the mathematical model. Mass transfer calculated using Dincer and Dost analytical approach showed Deff and hm in the range of 1.980 × 10−9 to 1.839 × 10−8 m2/s and 1.881 × 10−6 to 5.653 × 10−6 m/s, respectively. Regardless of puree thickness, samples dried at 70 °C exhibited superior quality, followed by those dried at 80 °C and 90 °C. The 2-mm-thick puree, dried at 70 °C, displayed the highest antioxidant activity (82.893