Prediction of the Physicochemical Properties of Spray-Dried Black Mulberry (Morus nigra) Juice using Artificial Neural Networks

An artificial neural network (ANN) was developed to predict quality changes of spray-dried black mulberry (Morus nigra) powder. In this study, the effects of inlet-air temperature (110, 130, and 150 °C), compressed air flow rate (400, 600, and 800 L/h), and concentration of different carrier types such as 6, 9, and 20 dextrose equivalent maltodextrins, and Arabic gum (8, 12, and 16%), were studied on five performance indices, namely: drying yield, bulk density, color change, total anthocyanin content, and antioxidant activity. A feed-forward multi-layered perceptron trained by back propagation algorithms for six independent variables was developed to predict the five outputs. The (6:14:5)-multi-layered perceptron, namely, a three-layered network having fourteen neurons in the hidden layer resulted in the best-suited model estimating the outputs at all drying runs. Predictive ability of ANN was compared using a separate dataset of 48 unseen experiments based on root mean square error, mean absolute error, and coefficient of determination for each output parameter. The optimum model was able to predict the five output parameters with coefficient of determination higher than 0.905. The results indicated that the experimental and ANN predicted data sets were in good agreement, so it is feasible to use ANN to investigate and predict the properties of black mulberry juice powder.