Artificial neural network modeling of microwave-assisted heat pump drying process

Abstract

Recently, microwave-assisted drying is an emerging drying technique that can help to accelerate the water removal process. In this paper, a microwave-assisted experimental drying study of carrot slides is presented. A microwave-assisted heat pump dryer is fabricated where the drying agent temperature, the supplied microwave power, and the circulative fan speed are easy to adjust. A series of 61 drying experiments is conducted with a range of drying temperature from 35 °C to 45 °C, a range of microwave power from 0 to 1.25 W/gram, and air velocity varied from 0.55 m/s to 1.70 m/s. The drying kinetics of the dehydration process is analyzed. Based on the obtained result, suitable intensive drying conditions for carrot slices are proposed. To extend the applicability of the experimental study, an artificial neural network (ANN) model of the drying process is developed. Instead of estimating the network for individual drying conditions in previous studies, the network is established for the entire range of drying conditions. The results indicate that the proposed ANN model predicts the microwave-assisted drying process adequately.