Application of Artificial Intelligence to Predict energy and exergy Efficiency of Convection Dryer for kiwi Slices

Abstract

This study aims to investigate and analyses the energy aspects of a hybrid convection-microwave drying method for kiwi slice. Several factors, including air temperature (50, 60, and 70 °C), air velocity (0.75, 1.5, and 2.25 m/s) and thickness (2, 4 and 6 mm) were evaluated to optimize the responses of energy efficiency (%), and exergy efficiency (%). An Artificial Neural Network (ANN) model was employed to model and optimize the aforementioned drying parameters.  Interestingly, the findings showed that increasing the air temperature led to higher energy and exergy efficiency, while higher thickness rates and air velocity resulted in lower energy and exergy efficiency.  This study has significant potential to enhance the efficiency of the kiwi slice drying process and contribute to energy conservation. The ANN model can be considered a powerful tool for predicting and determining optimal drying conditions, including energy and exergy efficiency. Based on the analysis of the modeling data using artificial neural network software, a network with a structure of 3-10-10-1 and 3-11-10-1 was selected as the most suitable network for estimating energy efficiency and exergy efficiency, respectively. The value of the R2 for predicting energy efficiency and exergy was calculated by the network as 0.9901 and 0.9869, respectively.