Modelagem e Controle da Separação de Enantiômeros em Leito Móvel Simulado

• Autor(a) principal: Tavares, Stella Ramos
• Data de Publicação: 2018
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFU
• Texto Completo: https://repositorio.ufu.br/handle/123456789/25698; http://dx.doi.org/10.14393/ufu.di.2019.315
• topic: Leito Móvel Simulado; Enantiômeros; Separação; Controle Predititvo; Simulated Moving Bed; Enantiomers; Separation; Predictive Control; Engenharia química; Enantiômeros - Separação; Controle Predititvo; CNPQ::ENGENHARIAS::ENGENHARIA QUIMICA

The industrial importance of the separation of enantiomers is not recent. However, the interest of the pharmaceutical industry in the use of this technology has been growing in the last decade. Chromatographic separation tests made with Simulated Moving Bed (SMB) have operational advantages compared to competitive technologies but adds complexity to the system because their own hybrid nature and its computational nature demands to make an excellent choice in the decision making process. The objective of this dissertation is to evaluate the behavior of an SMB structure and to investigate the configurations and changes that can be used for automatic control of a plant in which there are inherent complexities due to its intrinsic nature of non-stationarity, large time delay and disturbance sensitivity. Through the modeling of the system, simulations were made to better understand the process. As it was possible to see the occurrence of the cyclic steady state in both compounds to be analyzed, which will be denominated by A and B, with A being withdrawn at the raffinate output and B is the extract output. The component A separation achieves an efficiency of approximately 94%, and the component B, being more retained in the solid, around 81%. Changes were made in the input and output streams for process response analysis. When increased by 20% the output flow rates and also decrease at the same rate the input stream, it was possible to reach the cyclic steady state for both components. A novel model based predictive control strategy was introduced and gave satisfactory results for controlling a case study for praziquantel.