A Study of the Statistical Discrimination of the Drying Kinetics Equations

The drying kinetics of soybean seeds in a thin layer were studied in this work. In the thin layer drying experiments, air of constant humidity and a range of temperature and mass flowrate conditions, flowed through a thin layer of the wet material. The drying behaviour of the material was observed over a period of time. Most kinetic equations presented in the literature are nonlinear, so care should be taken when estimating parameters, since in some situations the estimators may not be appropriate. There are some procedures available to validate the statistical properties of the least squares estimators of nonlinear models. In this paper, the most important semi-empirical drying kinetics equations are considered and, by examining a experimental data set, through nonlinearity measures, the best equation is identified. The results for curvature measures, bias measures and simulation studies showed that the only equation that gives good inference results, based on least squares estimators, is the Overhults equation.