Estimação de Estados de uma Câmara Termoeletricamente Controlada utilizando Projetos de Filtros H1

2020 ◽  
Author(s):  
Bárbara J. de Oliveira ◽  
Víctor C. S. Campos ◽  
Márcio F. Braga
Keyword(s):  

O objetivo deste trabalho é apresentar e comparar duas estratégias de filtros para estimação de estados aplicados em uma Câmara Termoeletricamente Controlada (CTC). A CTC é composta por cinco sensores digitais de temperatura que representam os estados do sistema. A aplicação dos métodos é executada em duas etapas. Na primeira etapa, os filtros são implementados de forma off-line no software MATLAB R2018a, a partir dos dados reais do sistema. Enquanto, na segunda etapa, os filtros são aplicados em tempo real no sistema físico. São apresentados dois teoremas para a obtenção dos parâmetros dos filtros robustos tendo como base o lema de Finsler e o bounded real lemma, utilizando a norma Hinf como critério de desempenho, uma vez que emprega-se um modelo politópico incerto para descrever a dinâmica da CTC. O problema de filtragem ótima para o sistema é resolvido por meio de Desigualdades Matriciais Lineares. Os resultados obtidos pelos dois métodos são comparados graficamente e por meio da métrica do Erro Quadrático Médio.

2014 ◽  
Vol 525 ◽  
pp. 646-652
Author(s):  
Min Bian ◽  
Qing Yun Guo

The robust H2/<em>H</em>∞ control strategy for a class of linear continuous-time uncertain systems with randomly jumping parameters is investigated. The transition of the jumping parameters is decided by a finite-state Markov process. The uncertainties are supposed to be norm-bounded. It is desired to design a linear state feedback control strategies such that the closed-loop system satisfies H performance and minimizes the H2 norm of the system. A sufficient condition is first established on the existence of the robust H2/<em>H</em>∞controller bases on the bounded real lemma. Then the corresponding state-feedback law is given in terms of a set of linear matrix inequalities (LMIs). It is showed that this condition is equivalent to the feasible solutions problem of LMI. Furthermore, the control strategy design problem is converted into a convex optimization problem subject to LMI constraints, which can be easily solved by standard numerical software.


2012 ◽  
Vol 43 (11) ◽  
pp. 2062-2071 ◽  
Author(s):  
Deyuan Meng ◽  
Yingmin Jia ◽  
Junping Du ◽  
Fashan Yu
Keyword(s):  

Author(s):  
Roberd Saragih ◽  
Yoshida Kazuo

Abstract In this paper, we propose an order reduction method of controller based on combination of the alternating projection method and the balanced truncation. In this method both the errors of controller and the closed-loop system caused by the reduced-order controller can be improved simultaneously. By using a generalized Bounded Real Lemma, a feasible reduced-order controller can be derived. The sufficient condition for the existence of a reduced-order controller leads to a non-convex feasibility problem. To solve the problem, we can use an improved computational scheme based on the alternating projection method. But it is needed so much time to solve the problem if compared by the other methods. To validate the proposed method, some numerical calculations and simulations are carried out.


2017 ◽  
Vol 2017 ◽  
pp. 1-8
Author(s):  
Ting Hou ◽  
Jie Wang ◽  
Yueying Liu ◽  
Xiao Shen

With the help of a stochastic bounded real lemma, we deal with finite horizon H2/H∞ control problem for discrete-time MJLS, whose Markov chain takes values in an infinite set. Besides, a unified control design for H2, H∞, and H2/H∞ is given.


2011 ◽  
Vol 44 (1) ◽  
pp. 2332-2337 ◽  
Author(s):  
Michael M. Tchaikovsky ◽  
Alexander P. Kurdyukov ◽  
Victor N. Timin

Sign in / Sign up

Export Citation Format

Share Document