Sensitivity analysis of a mathematical model to simulate aeration of wheat stored in Brazil

Abstract

Abstract The research objective was to determine the relative importance of several variables in a non-equilibrium mathematical model for simulating aeration of stored wheat with uniform and non-uniform airflow distributions. The relative importance of each variable was determined by adding or subtracting fixed uncertainties from that variable and by calculating the effect of these changes on the predicted grain deterioration, after 1 year of storage. Results of simulations using weather data from Curitiba, Brazil, ventilation time of 06:00–12:00, fan temperature rises of 1 and 3 °C, and linear airflows of 0.0056 and 0.0278 m3/s per m2 indicated that the most important variables, in decreasing order, were the fan temperature rise, thin-layer wetting equation, and thin-layer drying equation. Wheat bulk density can be a constant and the net heat of sorption can be neglected. The ratio of bin diameter to bin height and the accuracy of the equations to describe wheat specific heat, air resistance to airflow, and equilibrium moisture content for adsorption and desorption were not important in the mathematical model. The deterioration model must be improved because the uncertainty in the calculation of wheat deterioration was much higher than the uncertainty generated by many of the other variables for the Brazilian climate.