The use of nonlinearity measures to discriminate the equilibrium moisture equations for Bixa orellana seeds

In this work, equilibrium moisture data for Bixa orellana seeds at different temperatures and relative humidities were determined using the Thermoconstanter Novasina instrument. Since the majority of the sorption equilibrium equations presented in the literature are nonlinear, care should be taken in estimating their parameters, because in some situations, the least squares estimators may not be appropriate. Therefore, some procedures are available in the literature to validate the statistical properties of the least squares estimators of nonlinear models. In this work, the nonlinear measures are used to discriminate between five model equations that represent the sorption equilibrium isotherms of Bixa orellana seeds. Results show that the Halsey modified equation is the best one to describe the experimental data because it has nonsignificance bias and nonlinear measurement parameters.