Neurogenetic Modeling of Moisture Sorption Isotherms in Dried Acid Casein

Abstract

A hybrid computational neurogenetic modeling (CNGM) algorithm has been investigated to predict moisture sorption isotherms in dried acid casein powder at three temperatures, i.e., 25, 35 and 45 degrees centigrade, and over water activity range of 0.11-0.97. The neurogenetic model was developed using a novel algorithm, which was utilized for training neural network rather than traditional learning methods like error back-propagation method. Also, six conventional empirical models, viz., Oswin, Smith, Halsey, Caurie, modified Mizrahi and Guggenheim-Anderson-de Boer (GAB) models were considered from elsewhere (that were fitted to the same data as used in this study) for comparison of the neurogenetic models' prediction potential. Accordingly, neurogenetic and GAB (best among the conventional models studied) models predicted sorption isotherms with accuracy, in terms of root mean squared percent error, ranging as 0.18-0.26 and 1.93-5.78 for adsorption; and 0.17-0.39 and 1.40-5.01 for desorption, respectively. Evidently, neurogenetic models outperformed conventional empirical sorption models. Hence, it is deduced that hybrid CNGM approach is potentially intelligent precision modeling tool for predicting adsorption and desorption isotherms in dried acid casein powder.