• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 1
  • 1
  • Tagged with
  • 2
  • 2
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Temperature Control of Multi-Product Semi-batch Polymerization Reactors

Clarke-Pringle, Tracy 07 1900 (has links)
<p> The work in this thesis focuses on the temperature control of a semi-batch polymerization reactor. The system is published by Chylla and Haase (1993) as an Industrial Challenge and is typical of reactors at S.C. Johnson Wax. The challenge is to find a single controller that can adequately regulate reactor temperature despite changing process conditions. The multi-product nature of the system makes it a particularly interesting problem. Several different controllers are implemented and evaluated in this thesis. The controllers are in part chosen to quantify the amount process information (large or small) required in a controller structure in order to achieve satisfactory control. Two of the most promising controllers are a PID with feedforward compensation and a Nonlinear Adaptive algorithm. It is found that in many cases, there may be little incentive to go to a complex model based controller as the simpler feedback algorithm provides adequate control. However, the nonlinear adaptive controller is more easily extended to multi-batch or multi-product situations because of its more general nature. The PID with feedforward compensation requires retuning for each new situation in order to maintain satisfactory control. </p> / Thesis / Master of Engineering (MEngr)
2

Otimização dinamica de um reator de polimerização pela aplicação da metodologia IDP / Dynamic optimization of a batch polymerization reactor by the apllication of IDP methodology

Pereira, Paulo Roberto Alves 13 August 2018 (has links)
Orientador: Ana Maria Frattini Fileti / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Quimica / Made available in DSpace on 2018-08-13T11:28:04Z (GMT). No. of bitstreams: 1 Pereira_PauloRobertoAlves_D.pdf: 4037300 bytes, checksum: da84be7f1c008bf7d365f44936f53414 (MD5) Previous issue date: 2007 / Resumo: Um grande número de indústrias químicas, bioquímicas e farmacêuticas, utiliza reatores de polimerização em batelada para a produção de polímeros especiais de alto valor agregado. Para tanto, são requeridas ferramentas de controle que forneçam informações confiáveis e otimizem as principais variáveis envolvidas na reação, conferindo propriedades específicas aos polímeros. Para o presente trabalho, adotou-se a reação de polimerização em batelada do metacrilato de metila (MMA) via radicais livres em solução, formando o polimetacrilato de metila (PMMA). O peróxido de benzoíla foi usado como iniciador e o acetato de etila como solvente. Para as etapas da reação do MMA, utilizou-se um modelo matemático fenomenológico conhecido e testado experimentalmente. Para o controle ótimo da temperatura do reator (T), manipulou-se a vazão do líquido de resfriamento (Fcw) e a potência de aquecimento on/off (Q) para o reator em batelada, composto de vidro encamisado, com capacidade para um litro de mistura reacional e um litro de fluido de resfriamento em sua jaqueta. Para a otimização do sistema MMA-PMMA, optou-se pela metodologia IDP (Iterative Dynamic Programming). Desenvolveu-se um programa IDP em Matlab, cuja confiabilidade foi testada com sucesso por um exemplo disponível na literatura especializada. Os resultados mostraram que o programa IDP é capaz de otimizar diferentes características de interesse como a conversão do monômero (Xm), representando um potencial de melhoria econômica. A comparação dos resultados das trajetórias de T e de Fcw obtidos pela otimização off line via IDP, com os resultados experimentais obtidos por Nunes (2004) e Antunes et al. (2005), apresenta comportamentos discrepantes ao longo do tempo. Esta discrepância pode explicar o rendimento experimental de 0,4 para Xm, cujo setpoint para T foi fixado em 63 ºC (336 K) e o seu controle executado pelo modelo PIDFuzzy, contra 0,6 da otimização off line via IDP, que considera os valores de Xm obtidos em cada estágio P para indicar novos setpoints de T. Logo, pode-se inferir que não é o controle de T através de um setpoint fixo que irá permitir um valor máximo de Xm, mas o controle de T para setpoints variáveis em cada estágio de tempo P, definidos pela trajetória de Xm obtida na otimização off line. Esta constatação pode ser considerada a grande contribuição da metodologia IDP para este trabalho. Portanto, pode-se afirmar que o programa IDP desenvolvido é adequado para a otimização off line do processo de polimerização do MMA. / Abstract: A considerable number of chemical, biochemical and pharmaceutical industries, make use of batch polymerization reactors in their processes, mainly to produce polymers with special characteristics and high aggregated value. In order to supply products with high quality standards required by the market, control tools are necessary to provide as fast as possible reliable processes information and at the same time, to optimize the main variables involved in the reaction and attribute special properties to the polymers. This work is concerned with the methyl metacrilate (MMA) free-radical batch solution polymerization reaction, which produces the polymethyl methacrilate (PMMA). Ethyl acetate is used as solvent and benzoyl peroxide as the reaction initiator. To represent the MMA polymerization reaction, a known and experimented phenomenological mathematical model was used. For the reaction optimal temperature (T) control were defined as manipulated variables the inlet cooling water flow rate (Fcw) and the heating power (Q). The equipment is a glass reactor of one liter capacity surrounded by a glass jacket for the heat exchange with the cooling water. To optimize the MMA-PMMA system, a program based on IDP (Iterative Dynamic Programming) technique was developed in Matlab and successfully tested through an example from the specialized literature. The results showed that the IDP program is capable of optimizing different characteristics of interest as the monomer conversion (Xm), and may represent a potential of economic improvement. The comparison between the results of trajectories T and Fcw from the off line optimization via IDP, with the experimental results presented by Nunes (2004) and Antunes et al. (2005), shows discrepant behaviors along the reaction time. This discrepancy may explain the poor experimental value of 0.4 for Xm, once the set point for T was set up in 336 K and its control performed by a PID-Fuzzy model during the runs, against a value of 0.6 for Xm from the off line optimization trough IDP program, which suggests new set points based on the results of Xm for each stage P. Therefore, it is possible to affirm that the use of a fixed set point to control T during the reaction probably will not lead the reaction to a maximum value of Xm, but the use of variable set points for each time stage P, according to off line optimization results for Xm. This description may be considered the major contribution from the IDP methodology for this work. Therefore, the program IDP developed in this work can be considered suitable for the off line optimization of MMA polymerization reaction. / Doutorado / Sistemas de Processos Quimicos e Informatica / Doutor em Engenharia Química

Page generated in 0.1145 seconds