Controle preditivo distribuído de processos lineares

• Ano de defesa: 2014
• Autor(a) principal: Rocha, Rosiane Ribeiro
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; BR; Programa de Pós-graduação em Engenharia Química; Engenharias; UFU
• Palavras-chave em Português: Controle preditivo baseado em modelo; Controle distribuído; Decomposição de modelos; Controle preditivo; Distributed control; Model predictive control; Decomposition models; CNPQ::ENGENHARIAS::ENGENHARIA QUIMICA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/15254; https://doi.org/10.14393/ufu.di.2014.36

The main purpose of a plant control is to coordinate the various interactions between the subsystems that compose it. The subsystems of a plant are usually designed independently or added later with the evolution of the installed plant. These changes usually occur motivated by production requirements or environmental regulations. Most large-scale systems implement the decentralized control as control strategy. But for subsystems with strong interactions, this approach can lead to unacceptable performance. Furthermore, centralized control is able to address optimally the problem of interaction, but with high structural and organizational costs, making costly such a complex structure and upgrade maintenance. A structure that preserves the topology and exibility of decentralized control and at the same time may offer a nominal closed-loop stability guarantee is the distributed control approach. In this control structure, the interactions between subsystems are modeled and information between the subsystems is shared between them. In this work , the plant model decomposition is proposed directly on the state space (A, B and C) matrices realization description, which has interactions between states, between states and inputs and between states and outputs. Each new subsystem has a subset of inputs and the selection of the states belonging to this subsystem is made through the analysis of the states that are excited by the subset of input selected for each subsystem. A decomposition algorithm is introduced that gives a set of controllable subsystems using a subset of the available inputs. The proposed DMPC was successfully evaluated in 4 studies of different cases and their performances compared with other control strategies (centralized and decentralized). The sequential and parallel structures were also evaluated and cases that the control problem with constrains on the manipulated variables was considered. The results show that the proposed DMPC presents performance satisfactory for the DMPC sequential and the order of MPC controllers is an important parameter in the design of this structure, for the first three examples. The parallel structure did not favor the performance of the proposed DMPC, except for the case study # 4, in which there are no differences between sequential and parallel structures.