Comparison between Artificial Neural Network model and mathematical models for drying kinetics of osmotically dehydrated and fresh figs under open sun drying

Abstract

In this study, drying kinetics of figs (Ficus carica L) under open sun drying (OSD) was compared with using Artificial Neural Network (ANN) model and mathematical models. Pulsed vacuum osmotic dehydration (PVOD) was applied as a pretreatment. In experiments, whole figs (Sarilop variety) were immersed in sucrose solution at 50 degrees Brix and 50 degrees C with a ratio of solution/food was 4:1 for 180 min. Vacuum impregnation was applied at 253 mbar for 15 min. Shrinkage effect was considered, also. Results showed that PVOD reduced drying time of figs. The mean of D-eff values of PVOD treated and fresh figs were obtained as 2.55 x 10(-10) m(2)/s and 2.36 x 10(-10) m(2)/s, respectively. It was found that present model was the best drying model with the highest correlation coefficient (R-2) values among the mathematical models whereas ANN model had a higher correlation coefficient (R-2 = 0.9999) values than the mathematical models for both PVOD treated and fresh figs.