Controle preditivo sintonizado via algoritmo genético aplicado em processos siderúrgicos

• Ano de defesa: 2011
• Autor(a) principal: Almeida, Gustavo Maia de
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Espírito Santo; BR; Doutorado em Engenharia Elétrica; Centro Tecnológico; UFES; Programa de Pós-Graduação em Engenharia Elétrica
• Palavras-chave em Português: Otimização matemática; Algorítmos genéticos; Identificação de sistemas; Controle preditivo; Medidas Elétricas, Magnéticas e Eletrônicas; Instrumentação; 621.3
• Link de acesso: http://repositorio.ufes.br/handle/10/9710

Techniques of Model Based Predictive Control (MPC) are increasingly applied in industry because they generally exhibit a good performance and robustness, since the parameters of controller are tuned correctly. This thesis use the Genetic Algorithm (GA) to perform the tuning of parameters of the predictive controller to control mono and multivariable, linear and nonlinear models. The existing technical literature for tuning Predictive Dynamic Matrix Controller (DMC), which use a step response to control systems that are open loop stable will be compared with the tuning by GA. In the event that the process is unstable in open-loop, there is not an analytical method for the tuning. Therefore, it is noted in the literature that the use of MPC in some open-loop unstable systems, linear or not (Such as these to be studied in this thesis) is lacking, because the tuning procedure is based on trial and error and sometimes is impractical. Therefore, this study focuses the application of the MPC tuning by Genetic algorithm for two open-loop unstable processes, which are very important in the Steel Industry. The first is composed by Rolling Mill Stands, where we wish to minimize the variation of strip thickness the last stand due to disturbances that affect the process such as temperature and strip thickness variations in the first stand. In this case we use the linear and multivariable model to develop the Generalized Predictive Controller (GPC) whose parameters are tuned by GA. The second process, unstable in open-loop, is the level of Mold of a Continuous Casting, which has a nonlinear model and is therefore controlled by techniques of nonlinear predictive control using neural networks and Hammerstein model. A comparison is made between these controllers to analyze the stability and robustness when the mold is affected by disturbance of Bulging, Clogging and Argon.