Proposition d'une approche intégrée basée sur les réseaux de Petri de haut niveau pour simuler et évaluer les systèmes contrôlés en réseau
Belynda Brahimi
- Year
- 2007
- Citations
- 3
- Access
- Open access
Abstract
The Networked Control Systems (NCS) used in collaborative, and distributed applications are based both on identification of application requirements named Quality of Control (QoC), and on performance evaluation of network to obtain the required Quality of Service (QoS). The pluridisciplinary aspect of NCS needs the knowledge of many domains such as information theory, robotics, sensors and networks. In general, two approaches are investigated in NCS: the first one deals with the compensation of the network perturbations by using control theory (control over network). The second one adapts the network performances according to the application needs (control of network). The objective of this thesis is to propose an integrated modelling environment to represent globally the NCS behaviour by using the High Level Petri Nets (HLPN) formalism. This work considers NCS based on switched Ethernet architectures. Such communication architecture is modelled with HLPN and is evaluated according to different scheduling mechanisms and traffic load. After that, NCS model based on HLPN is proposed. This model integrates the Ethernet switch model and applicative environment of NCS: controller, process... Finally, the strategies to control the network in order to adapt the QoS according to the QoC required by the application are proposed. These strategies are achieved using priority static and Weighted Round Robin policies. The obtained results show that the scheduling mechanisms enable to improve the performance of communication system and then to improve the application performances.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Fractional Differential Equations
Igor Podlubný
2025
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991
Genetic Programming: On the Programming of Computers by Means of Natural Selection
John R. Koza
1992