Secagem do milho em camada espessa: análise experimental e simulação numérica tridimensional em secador cilíndrico de leito fixo com admissão radial do ar de secagem

Abstract

Brazil is the third largest corn producer in the world and the second in exports, in addition to presenting a growing volume of cereal harvests due to the introduction of trade with Asian countries. In this way, the need for drying the grains is increasing, aiming at adequate and long-term processing. Thus, all the improvement and monitoring of the corn grain drying process provide a significant benefit for small and large producers. In this work, a validated three-dimensional model was obtained for drying corn grains of the BG 7640 VYH variety in a cylindrical dryer with radial air admission, using equations for the thin layer and physical properties adjusted for the cultivar itself, and models with these characteristics were not found in the literature. For this purpose, mathematical models of thin layer grain drying were tested and adjusted for corn drying conditions. A laboratory-scale drying column was developed to analyze thick layer drying, based on a large cylindrical radial dryer. Experimental tests were carried out aiming at the engineering of the equilibrium temperature in the grains at certain air temperatures. Thermocouples positioned inside the bed were used to monitor grain temperature. Static pressure and air flow velocity at the inlet were monitored. To experimentally test how the thin layer models were used, a drying column of smaller dimensions was used. Six different thin-layer models were tested, and after adjustment the Page, Midilli and Overhults models stood out, with coefficients of determination (R2) equal to one and relative errors lower (P) than 1%. Thick layer drying was carried out in the radial cylindrical drying column constructed, for the speed of 2 and 4 m.s-1 at temperatures of 40, 50 and 60ºC, the tests provided experimental data for validation of the mathematical model. The simulation was performed using the COMSOL Multiphysics software, which uses the finite element method to solve differential equations. The numerical results for the temporal variation of the average moisture content of the bed, local moisture content of the corn and air temperature were compared with the experimental data. All data for moisture content corroborated providing values greater than 0.97 for R2 and less than 5% for P. For air temperature, P values between 4 and 15% were obtained. After validation, simulations were performed for a medium-sized dryer, which showed energy and exergy efficiencies of 45 and 5%, respectively.