Generalization of a lumped parameters model using fractional derivatives applied to rice hydration

This study presents a generalization of a lumped parameters model through the use of fractional derivatives of arbitrary order. The conventional model has an analytical solution in terms of exponential function. However, when applied to kinetic data on moisture absorption in rice grains, its performance was worse when compared to the proposed generalized model. The results were analyzed for four hydration temperatures. Despite the similarity of the mass transfer coefficients for both models, the proposed generalization provided a statistically significant improvement in prediction of moisture values in relation to time and process temperature. Practical applicationsThe model proposed in this paper is applicable in describing the hydration kinetics of rice which is an important step in parboiling process, for instance. The simplicity of the analytical solution of the model and the improvement of its prediction capability makes the presented approach useful to project of equipment intended to process rice.