Evaluation of the Operational Parameter's Effect on Mass Transfer Kinetics During Osmosic Dehydration of Potato Slices Using Sensitivity Analysis Based on Artificial Neural Networks

Osmotic dehydration, as a minimal processing method, has found increasingly wide prospects during the past few decades. This process involves mass transfer which is commonly modeled by applications of different procedures, mostly based on Fick's law. In this research, we approached the modeling process by first obtaining experimental measurement of potato’s solid gain, water loss and moisture content under different conditions of solution concentrations (5, 10 and 15% w/w), temperatures (30, 40 and 60oC), potato to solution ratio (1:6, 1:8 and 1:10) as well as time intervals (1, 2, 3 & 4h). In order to evaluate the effect of changes in operational parameters on mass transfer kinetics, sensitivity analysis was performed. Artificial neural networks (ANN) was applied for modeling. The results exhibited how much powerful the model is in prediction of the system’s outputs, and high sensitiveness of these outputs to the ratio of potato to osmotic solution.