Programa de Pós-graduação
em Engenharia Elétrica


Defesa Pública do Exame de Qualificação para Doutorado do aluno GUSTAVO RODRIGUES GONÇALVES DA SILVA

Data: 10/11/2017 - sexta-feira
Horário: 09h00min
Local: Salão de Eventos do Instituto Eletrotécnico da UFRGS (Av. Osvaldo Aranha, 103 - 1º andar)

Orientador: Prof. Dr. Alexandre Sanfelici Bazanella - PPGEE - UFRGS

Banca examinadora:

Prof. Dr. Michel Gevers - Université Catholique de Louvain, Bélgica
Prof. Dr. Reinaldo Martinez Palhares - Departamento de Engenharia Elétrica - UFMG
Prof. Dr. Diego Eckhard - PPGEE - UFRGS
Prof. Dr. Jeferson Vieira Flores - PPGEE - UFRGS


"This dissertation addresses Data-Driven control methods for multivariable systems, with focus on Non-Minimum Phase systems and an approach for state-feedback in the Linear Quadratic Regulator framework. Since fundamental assumption in Data-Driven control is that there is no model of the plant available, these methods come with no guarantee to yield a stabilizing closed-loop, thus the controller certification problem is also tackled in a Data-Driven perspective. The goal is to achieve enhanced Multivariable Data-Driven control methods to cope with a variety of systems and that yield stabilizing controllers and performance similar to the one specified. One approach is the direct extension of existent Data-Driven methods, which deals with the Model Reference control problem therefore considering transfer matrix systems representation. In this case, special attention is given to Non-Minimum phase systems and in this work two Data-Driven methods are extended with a flexible criterion to cope with this issue and different structures of the reference model. Another option is to consider state-space systems representation and the corresponding state-feedback design. For this approach an algorithm was created to cope with the infinite horizon Linear Quadratic Regulator problem, based on a previous attempt to solve the problem; in this case there is no issue when dealing with Non-Minimum Phase systems. Finally, future contributions of this work are a purely Data-Driven estimation of the H∞-norm from above and study the conservativeness of metrics comparing two controllers. These tools are applied to the controller certification problem which will be used to try to solve the long-lasting stability guarantee issue in Data-Driven methods. Usage of these methods in simulation experiments and actual systems shows the applicability and improvement of the proposed enhancements. The guarantee of stability is still an open issue and is the topic of future research. Therefore this work provides new tools and a certain level of polishing in the theory for multivariable Data-Driven control methods.
Keywords: Data-driven control, non-minimum phase systems, controller certification, multivariable control."