OATAO - Open Archive Toulouse Archive Ouverte Open Access Week

Supervision des régulateurs par logique floue

Ewers, Bernd and Bordeneuve-Guibé, Joël Supervision des régulateurs par logique floue. (1997) Revue française de mécanique (n° 2). pp. 135-141. ISSN 0373-6601

[img] (Document in English)

PDF (Publisher's version) - Depositor and staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
719kB

Abstract

Classical control theory is based upon an analytic system design procedure using deterministic or stochastic approaches. In the past years, there has been a growing interest in using new computational and machine intelligence technologies such as rule-based systems, fuzzy logic or neural networks in process control. These approaches are often referred to as expert or qualitative control. Fuzzy logic, in particular, has been successfully applied to various industrial control problems. Generally speaking, fuzzy logic and rule-based techniques are a means of dealing with imprecision, a method of modeling human behavior, allowing to incorporate heuristic knowledge and symbolic information into process calculators. This approach becomes particularly interesting when a sufficient analytic representation of the process is too difficult or even impossible to obtain. However, although expert control is more and more accepted in the control engineering community, its place among conventional techniques has not been defined clearly. This is primarily due to the lack of a general methodology to systematically validate the global functioning and performance of systems that incorporate intelligent components. Nevertheless, it can be observed that expert control and symbolic data processing have the potential to improve the performance of a feedback loop and to provide new functions in a control system. As a part of a research program to study the potential role of fuzzy logic for future aircraft control systems, this paper focuses on the conception of supervisory control structures for aeronautical applications using rule based expert systems and fuzzy logic theory. This concept can be used to backup or monitor a conventional control system improving its performance in the presence of a changing system environment or in extreme operating conditions. The effectiveness of this method could be successfully demonstrated by two application examples, which will be discussed in this paper.

Item Type:Article
Audience (journal):National peer-reviewed journal
Uncontrolled Keywords:
Institution: Université de Toulouse > Institut Supérieur de l'Aéronautique et de l'Espace - ISAE
Laboratory name:
Département de Mathématiques, Informatique, Automatique - DMIA (Toulouse, France) - Automatique, Dynamique et Interface des Systèmes - ADIS
Statistics:download
Deposited By: joel bordeneuve-guibe

Repository Staff Only: item control page