Artificial neural network model of pork meat cubes osmotic dehydration

Abstract

Mass transfer of pork meat cubes (M. triceps brachii), shaped as 1×1×1 cm3, during osmotic dehydration (OD) and under atmospheric pressure was investigated in this study. The effects of different parameters, such as concentration of sugar beet molasses (60-80 mass%), temperature (20-50 °C), and immersion time (1-5 h) in terms of water loss (WL), solid gain (SG), final dry matter content (DM), and water activity (aw), were investigated using experimental results. Five artificial neural network (ANN) models were developed for the prediction of WL, SG, DM, and aw in OD of pork meat cubes. These models were able to predict process outputs with coefficient of determination, r2, of 0.990 for SG, 0.985 for WL, 0.986 for aw, and 0.992 for DM compared to experimental measurements. The wide range of processing variables considered for the formulation of these models, and their easy implementation in a spreadsheet calculus make them very useful and practical for process design and control.