Modelling of sorghum soaking using artificial neural networks (MLP)

Introduction Artificial neural network is a technique with flexible mathematical structure which is capable of identifying complex non-linear relationship between input and output data. Objectives The aim of this study was evaluation of artificial neural network efficiency for simulating the soaking behaviour of sorghum kernel as a function of temperature and time. Methods In this study, soaking characteristics of sorghum kernel was studied at different temperatures (10, 20, 30, 40 and 50?degrees C) by measuring an increase in the mass of sorghum kernels with respect to time. A multilayer perceptron neural network was used to estimate the moisture ratio of sorghum kernel during 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. Results Results showed that the estimated moisture ratio by multilayer perceptron neural network is more accurate than Page's model. It was also found that moisture ratio decreased with increasing of soaking time and increased with increasing of soaking temperature. Conclusion The artificial neural network model was more suitable than other models for soaking behaviour estimation in sorghum kernel.