Simulação por fluidodinâmica computacional de um leito fluidizado na operação de secagem de minério de ferro

Abstract

Brazil is one of the leading exporters of iron ore, an important product due to its use in steel production and its significant impact on the trade balance. Given the high competitiveness in this sector, attention to ore quality and production costs is crucial. The selling price of iron ore is essentially related to its iron content and the moisture of the material. Brazilian ore from Carajás/PA is renowned for its high iron content. However, to increase the added value and comply with safe transportation standards, controlling its moisture is necessary. The ore can undergo a drying process to meet the moisture requirements for transportation and marketing. To optimize the drying process, it is crucial to analyze the dynamics of the involved multiphase flow. One of the main parameters for sizing a fluidized bed in drying operations is the minimum fluidization velocity (U_mf). In this context, the present undergraduate work aims to analyze, via Computational Fluid Dynamics (CFD) using the ANSYS Fluent software, the flow of iron ore in a bench-scale fluidized bed, using the Eulerian Granular multiphase model to describe the solid phase behavior. In particular, the study examines the influence of parameter variations such as the solid phase diameter on the minimum fluidization velocity and compares the simulated results with literature correlations. To achieve this, U_mf was obtained in CFD simulations for two solid phase diameters by analyzing the pressure drop behavior at the bed inlet for various fluid phase velocities. The simulation data revealed that the minimum fluidization velocity for a solid phase particle diameter of 0.001 m was 1.75 m/s, while for a diameter of 0.006 m, it was 6 m/s. It was observed that as the solid phase diameter increases, a higher velocity is required for fluidization because a larger particle diameter corresponds to a lower drag coefficient. Comparing with values obtained from numerical correlations, a deviation of between 18.8% and 58.2% was observed when using these correlations, as they are designed for specific cases. It was concluded that CFD is a tool with potential for estimating the minimum fluidization velocity in fluidized bed simulations.