Hossein Majidzadeh; Bagher Emadi; Abdolali Farzad
Abstract
In this study, the moisture content of kiwifruit in vacuum dryer was predicted usingartificial neural networks (ANN) method. The drying temperatures (50, 60 and 70ºC), vacuum pressures(500, 550 and 600 mmHg), thicknesses of kiwifruit slices (3, 5 and 7mm) and drying times were considered as the ...
Read More
In this study, the moisture content of kiwifruit in vacuum dryer was predicted usingartificial neural networks (ANN) method. The drying temperatures (50, 60 and 70ºC), vacuum pressures(500, 550 and 600 mmHg), thicknesses of kiwifruit slices (3, 5 and 7mm) and drying times were considered as the independent input parameters and moisture content as the dependentparameter. Experimental data obtained from vacuum drying process, were used for training and testing the network. Several criteria such as training algorithm, learning rate, momentum coefficient, number of hidden layers, number of neurons in each hidden layer and activation function were given to improve the performance of the ANN. The total number of hidden layers and the number of neurons in each hidden layer were chosen by trial and error. The best training algorithm was LM with the least MSE value. Optimum values of learning rate and momentum for the ANN with GDM training algorithm were set at 0.2 and 0.05, respectively. The optimal topologies were 4-20-1 with Tansig activation function and MSE values of 0.0016 and 4-15-20-1 with Logsig activation function in both hidden layer and MSE values of 0.000147. The correlation between the predicted and experimental values in the optimal topologies was higher than 99.75%.
