Modeling thin-layer drying of banana and process optimization using artificial neural network

One of the important processes to preserve agricultural crops after harvest is drying. Since drying time and condition has considerable impacts on the process economic efficiency and energy consumption, therefore, kinetic modeling of drying can be a suitable way for optimum process control. In this study, banana thin layer drying was performed at four air temperatures (50, 60, 70 and 80°C), three air velocities (0.5, 1 and 1.5 m/s) and three banana thicknesses (2, 4 and 6 mm). Results showed that the Page and modified Page models have the best fit with the experimental data. These models had higher R2 and lower RMSE and χ 2. Artificial neural network was also used to optimize thin layer drying of banana. The results demonstrated that the experimental data can be fitted well with the network (correlation coefficient: 0.9984).