Modelling and simulation of Ethernet based networked mechanical systems

• Main Author: Shaker, Vahid
• Format: Others
• Published: 2004
• Online Access: http://spectrum.library.concordia.ca/7835/1/MQ91111.pdf; Shaker, Vahid <http://spectrum.library.concordia.ca/view/creators/Shaker=3AVahid=3A=3A.html> (2004) Modelling and simulation of Ethernet based networked mechanical systems. Masters thesis, Concordia University.

Distributed control and simulation based on Ethernet networks has become increasingly common in recent years due to its superior performance and cost over other networking technology. In order to systematically design such distributed systems, which combine Ethernet network components and mechanical systems, a modular modelling and simulation approach for this class of systems is necessary. In this thesis, an innovative modular modelling and simulation approach is developed to predict the behaviour of distributed mechanical systems based on Ethernet (IEEE 802.3 protocol) networks. The main objective is to predict the overall time delay of transmitted packets and estimate the real-time performance of Ethernet/mechanical systems in different topologies for both normal and abnormal operating conditions. The approach is based on discrete finite state machine (FSM) models developed for the main types of Ethernet network components (NIC, bus, hub, and switch). The FSM models are implemented in C++ and encapsulated using Simulink S-functions. This allows the network models to be easily combined with Simulink mechanical system models that are common in control and simulation applications. The new approach is tested for a number of case studies to test its capability to predict time delays that arise from network communication. Finally, combined network/mechanical simulations are performed to illustrate the application of the new approach to distributed simulation problems. Together, these results provide a new approach for simulation of Ethernet based mechanical systems that can be used to design distributed control and simulation systems.