Predicting the Moisture Ratio of Dried Tomato Slices Uusing Artificial Neural Network and Genetic Aalgorithm Modeling

Abstract

Nowadays, mathemathical simulation and modeling of drying curves are useful instruments in order to improve control systems for final product quality under various conditions. These approaches are usually applied for studying the factors present in the process, optimization of the conditions and working factors as well as predicting the drying kinetics of products. Two intelligent tools including artificial neural network (ANN) and genetic algorithm (GA) were used in the current paper for predicting tomato drying kinetics. For this purpose, four mathematical models were taken from the literatures, then they were matched with the empirical data. Final step was choosing the best fitting model for tomato drying curves. According to the results, the model proposed by Aghbashlo et al (Agh-m) showed great performance in predicting the moisture ratio of the dried tomato slices. Moreover, the genetic algorithm was utilized for optimization of the best empirical model. Ultimately, the results were compared with the findings observed in ANN and GA models. The comparison indicated that the GA model offers higher accuracy for predicting the moisture ratio of dried tomato with the correlation coefficient (R2) of 0.9987.

Keywords

• artificial neural network
• genetic algorithm
• thin-layer drying
• tomato slice