A comparative study for the estimation of drying behavior of pomegranate arils: regression analysis and neural network

This paper presents drying kinetics of pomegranate arils and a comparative study between regression analysis and a multilayer feed-forward neural network to estimate its dynamic drying behavior. Experiments were performed at drying air temperatures of 40 50 , 60 and 70°c , with air flow velocities of 1 and 2 m/s in a convective dryer. Seven different mathematical models available in the literature were fitted to the experimental data. In addition , a three-layer feed-forward neural network was used to estimate the pomegranate arils dynamic drying behavior. A back propagation algorithm was developed ( using MATLAB ) and applied to training and the testing the network. Comparing the seven models and the feed-forward neural network , it was concluded that the neural network represented the drying characteristics better than the mathematical models.

Keywords:

Artificial neural network , Mathematical model , Moisture ratio , Drying , Neuron