Perediction of Moisture Ratio in Drying Process of Garlic Cloves by GMDH type-Neural Network

Drying kinetic of garlic cloves was investigated by considering different drying conditions. In this study, drying of garlic cloves was done by microwave-convective technique, using microwave powers of 100,180 and 300 W, air temperature of 40,100 and 140 0C. The samples sizes were about 1.09 gr. each, and with thickness of 5 and 7 mm and air velocity were held stable at 1m/s. A GMDH type-neural network was employed to estimate moisture content of garlic slices. Accuracy of the models was measured using the coefficient of correlation (R2) and root mean square error (RMSE). ). The results were found that GMDH had a good agriment with experimental data

Keywords:

Drying , Garlic Cloves , Microvawe-Convection , Pareto , GMDH Type-Neural Network