Modelagem, simulação e controle de colunas de destilação apropriadas à produção de etanol em pequena escala

• Autor(a) principal: Ketzer, Felipe
• Data de Publicação: 2013
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Manancial - Repositório Digital da UFSM
• Texto Completo: http://repositorio.ufsm.br/handle/1/7963
• topic: Etanol combustível; Destilação; Mini usinas; EMSO; Fuel ethanol; Distillation; Small-scale plant; Process control; Controle de processos; CNPQ::ENGENHARIAS::ENGENHARIA DE PRODUCAO

The environmental liabilities that are faced by fossil fuel industry motivate the search for renewable energy sources. Although Brazil is the second largest world producer of bioethanol, the current production deficit is expected to go up by 2015. In this direction, it is utmost important to widen the ethanol feedstock cultivated area, as well as expanding the number of production units of this biofuel. Small ethanol plants emerge as an interesting alternative for the decentralization of production. However, the design and operation of such plants require high efficiency to enable their operation. This work presents the study of modeling, simulation and control of a batch distillation column for the separation of binary mixture ethanol-water to produce hydrous ethanol fuel with a concentration between 95.1% (v/v) and 96% (v/v), according to ANP specifications. The distillation column, located at the Polytechnic College of the Federal University of Santa Maria, was used as basis for the fundamental modeling which was performed using the software EMSO. Different models of phase equilibria were analyzed, being the Van Laar model for calculating the activity coefficient of the mixture in the liquid phase, coupled with ideal vapor mixture model, the one that have shown the best results. In this study, different control strategies were also proposed and evaluated to keep the top product concentration on the ANP specification for hydrous ethanol fuel. The chosen control algorithm was the Proportional Integral designed with discrete linear parametric models and through different tuning methods found in the literature. Robustness and performance criteria were used to select the best controller. Three process variables, candidates to be the manipulated variable of the control strategy, were compared: heating steam flow rate of reboiler, cooling water flow rate of pre-condenser and reflux ratio. The results have shown that the most efficient strategy was the one in which the flow of cooling water flow rate of pre-condenser was used as the manipulated variable. Besides ensuring that concentration is kept quite close to the azeotrope formed by ethanol-water mixture, another advantage of this variable in a control strategy is to allow the production of a high hydrous ethanol fuel.