Desenvolvimento de peneira de alta frequência em escala de laboratório e validação de modelos matemáticos

• Ano de defesa: 2022
• Autor(a) principal: Matheus Naves Moraes
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; Brasil; ENG - DEPARTAMENTO DE ENGENHARIA METALÚRGICA; Programa de Pós-Graduação em Engenharia Metalúrgica, Materiais e de Minas; UFMG
• Palavras-chave em Português: Modelamento de processo; Peneiramento fino; Peneira de laboratório; Scale-up; Amostra reduzida; Engenharia de minas; Tecnologia mineral; Peneiramento (Mineração)
• Link de acesso: http://hdl.handle.net/1843/51130

The high frequency screening for fine classification is a process available for a few decades, but only with recent innovative upgrades allowed its adoption for mineral processing application. Throughout the last decade, high frequency screens started to be applied to grinding circuits, replacing or supplementing hydrocyclone’s classification with productivity benefits published in several studies. However, the trials carried out to validate its adoption were performed directly in a real scale industrial device. On one hand, a real scale trial leads to a higher accuracy for modeling and simulation. On the other hand, samples of a couple of tons are necessary and a robust installation is demanded to run the trials, which limits the screen assessment for new projects, especially greenfield ones. Therefore, this study describes the development of a laboratory scale high frequency screen for continuous trials, assessing selected screen models and their implications for simulation. In addition to the successful trials at the developed laboratory screen, it was observed that Mwale’s model is the most suitable for this process, since it has a good balance between accurate modeling and simulation capacity. Batterham et al has good accuracy for modeling, but unique solution, being unfeasible for simulation. Hatch&Mular and Rogers have good accuracy for modeling and are simple to be applied, however still with limitations for simulation.