Estudo da cinética de secagem de semente de abóbora kabocha (Tetsukabuto) utilizando a ferramenta MATLAB

Abstract

The present work, considering the importance of utilizing residual parts of fruits and vegetables, highlights drying as a method for preserving pumpkin seeds, which are typically discarded by the industry and consumers. Drying is one of the oldest preservation methods known in the literature, aiming to reduce the amount of free water in the material or food, preserve quality, and extend shelf life. The optimization of this process through mathematical modeling can be achieved through computational simulations, using empirical drying models with a direct relationship between moisture content and drying time. The use of MATLAB r2022a software facilitates the adjustment of empirical parameters to a mathematical model for the drying process. In this context, the present work aimed to study the drying of butternut squash pumpkin seeds in a forced-air circulation greenhouse at a speed of 2 m.s-1 and temperatures of 50, 60, and 70 ºC, and analyze the drying process using MATLAB r2022a to determine the most suitable mathematical model. The seeds were distributed in a single layer on an aluminum container, with periodic mass measurements until reaching constant mass. The models used to fit the drying curves were the Henderson and Pabis model, Midilli model, Newton model, and the Page model. The activation energy was calculated to compare the water diffusivity in the product under the study conditions, according to the Arrhenius equation. The calculated activation energy was 20.56 kJ mol-1, consistent with literature expectations ranging from 12.7 to 110 kJ mol-1 for agricultural products and close to other studies on drying fruit and vegetable residues. According to curve fitting parameters, mean relative error, and mean estimated error, the Midilli model demonstrated the best fit to the experimental data, especially at 50 ºC, with a low MRE of 0.0977, SE of 0.019, and X² of 3.61 x 10-4, indicating a good fit to the experimental data.