Controle supervisório com estruturas concorrentes de refinamentos sujeitas à dependência de contexto
The Supervisory Control Theory (SCT) defines the synthesis of controllers for Discrete Events Systems. Although within its 30 years of existence the SCT has been extended and improved in several ways, it still preserves a feature that limits its application on real industrial problems, that refers t...
| Autor principal: | Assmann, Maicon Paulo |
|---|---|
| Formato: | Trabalho de Conclusão de Curso (Graduação) |
| Idioma: | Português |
| Publicado em: |
Universidade Tecnológica Federal do Paraná
2020
|
| Assuntos: | |
| Acesso em linha: |
http://repositorio.utfpr.edu.br/jspui/handle/1/14597 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: |
The Supervisory Control Theory (SCT) defines the synthesis of controllers for Discrete Events Systems. Although within its 30 years of existence the SCT has been extended and improved in several ways, it still preserves a feature that limits its application on real industrial problems, that refers to the modeling complexity. In this sense, the idea of events refinement has been recently proposed in the literature as a way to simplify the modeling and to enable the application of the SCT over a wider range of problems. However, there are classes of processes over which the idea of refinement does not apply directly. For example, processes that include more than one layer of refinements in the same problem and control context. For these cases, the idea of refinement should be extended to allow and distinguish multiple sets of refinements characterized by context dependencies. This work proposes a formal approach that allows creating any number of refinements dependent from each other, which can be used, then, in the synthesis of controllers with event distinctions. The approach is illustrated using examples, in particular, a manufacturing system with rework and buffering of workpieces. In this example, is demonstrated a solution that controls in parallel the number of cycles and the buffer limits would be very hard to be modeled. Yet, using the proposed method, it is shown that the modeling can become quite simpler and intuitive. |
|---|