Modeling of wheat soaking using two artificial neural networks (MLP and RBF)

Abstract

In this study soaking characteristics of wheat kernel was studied at different temperatures (25, 35, 45, 55 and 65 °C) by measuring an increase in the mass of wheat kernels with respect to time. Artificial neural network (ANN) is a technique with flexible mathematical structure which is capable of identifying complex non-linear relationship between input and output data. A multi layer perceptron (MLP) neural network and radial basis function (RBF) network were used to estimate the moisture ratio of wheat kernel during soaking. ANNs were used to model wheat kernel soaking at different temperatures and a comparison was also made with the results obtained from Page’s model. The soaking temperature and time were used as input parameters and the moisture ratio was used as output parameter. The results were compared with experimental data and it was found that the estimated moisture ratio by multi layer perceptron neural network is more accurate than radial basis function network and Page’s model. It was also found that moisture ratio decreased with increasing of soaking time and increased with increasing of soaking temperature.