On the drag coefficient in particle agglomerates: a CFD approach to propose a new drag correlation

• Ano de defesa: 2023
• Autor(a) principal: Oliveira, Ricardo Arbach Fernandes de
• Orientador(a): Lopes, Gabriela Cantarelli
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Engenharia Química - PPGEQ
• Palavras-chave em Português: Material particulado; Aglomerados de partículas; Partículas de formatos irregulares; Coeficiente de arraste; Fluidodinâmica computacional
• Palavras-chave em Inglês: Particulate matter; Particle agglomerates; Irregularly-shaped particles; Drag coefficient; Computational fluid dynamics
• Área do conhecimento CNPq: ENGENHARIAS::ENGENHARIA QUIMICA::PROCESSOS INDUSTRIAIS DE ENGENHARIA QUIMICA
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/18583

In particle-laden flows, it is essential to model the fluid-particle interaction. However, depending on effects such as friction and attraction, surface properties, and collisions, particles can agglomerate, generating new irregularly-shaped particles. Such phenomenon is relevant since the geometry directly interferes with the flow dynamics. One way to evaluate how the particle will interfere is through the drag force. In the theoretical development of equipment, this force is taken into account and represented by the drag coefficient and is highly dependent on two variables: particle geometry and flow velocity. In the literature, several correlations are observed, obtained both in the experimental field or numerical field, using computational fluid dynamics (CFD). The advantage of using CFD lies in the ease of varying the flow velocity and in obtaining the drag coefficient from the pressure and velocity fields. However, most of these methods employ transient formulations, resulting in highly detailed but computationally expensive outcomes. This cost increases exponentially as the Reynolds number of the flow increases, making the study of turbulent flows infeasible. Consequently, these models are generally obtained for Reynolds numbers below 300 and then extrapolated to higher values when implemented in CFD codes. Thus, this study proposes an alternative approach to the problem by using steady formulation simulations, aiming to reduce computational costs and proposing a new, simple, and unified correlation for the drag coefficient capable of encompassing a wide range of flows, from laminar to turbulent, applicable to irregularly shaped particles and easily implementable in CFD codes. Using CFD simulations with experimental validation, it was possible to obtain the characteristic drag coefficient curve over a wide range of Reynolds numbers (0.1 ≤ Re ≤ 3500) for agglomerates of spheres representing irregularly-shaped particles. By individually simulating the flow around these agglomerates while varying the geometries and flow velocities, a new correlation for calculating the drag coefficient was proposed, capable of fitting the obtained characteristic curves. In general, the results showed that the use of a steady formulation can yield good results provided that the mesh is properly refined and the turbulence model accurately represents the flow. The new correlation, combined with the use of flatness as a geometric characterization parameter, proved effective in representing the drag curve, with maximum, minimum, and average deviations of 10.78%, -7.62%, and 3.79%, respectively, compared to simulated results, and 14.36%, -12.36%, and 9.6% compared to experimental results.