Abstract

Mohammed A. Al-Sulaiman

The ability of artificial neural networks to model non linear complex systems such as drying is increasing. The aim of this work is to develop an artificial neural network model, to predict quality indices of okra pods after drying in a domestic microwave oven that could be a tool used to predict some product quality. The optimal artificial neural network model was found to be a 3-6-4 structure with sigmoid transfer function. This optimal model was capable of predicting the total color change ( E), browning index, coefficient of rehydration and bulk shrinkage coefficient with R2 higher than 0.98 during training phase. It was concluded that the artificial neural network model predicted quality indices better than the multiple linear regression model.