Fuzzy Scheduling Strategy Based On Network Quality Of Service For Networked Control Systems

Networked Control Systems are a new control system integrated with control technologies, computer technologies and communication technologies. It can be used widely in control domain. With Integrated Development of the integration of Information Network and Control Network, the diversity of data has been increased obviously in the network. Then the problem of Networked Control Systems's optimization scheduling is increasingly outstanding. With the higher and higher requirement to the systems' comprehensive performance, the requirement of users can not be satisfied by existed scheduling schemes. Therefore, Fuzzy Logic Scheduling scheme based on Quality of Service is proposed in this thesis. As ensuring Quality of Performance of control systems, Quality of Service of aperiodic data is improved with Experimental Confirmation.The main work of this thesis as follows:(1) Basic properties of four popular network simulation tools, which are Truetime, NS-2, OPNET and Qualnet, are compared and analyzed. According to Networked Control Systems's newest research trend, a simulation experiment system of Networked Control System, which runs under Simulink, is designed and implemented with the Truetime kernel in this thesis. Many kinds of classical network scheduling schemes, as well as the conflict arbitration strategies, the network protocol types and the control algorithms are realized in this experiment system. System's properties can be analyzed online in this experiment system. Because the system is coded by Modular Design, then it can be re-developed by other researchers.(2) The characteristics of CAN net, Industry Ethernet and IEEE802.11e WLAN are studied by simulation. Advantages and disadvantages of Centralized Scheduling Scheme RM, MEF and Distributed Scheduling Scheme EDF, MTS are compared and analyzed.(3) Based on Quality of Service, Fuzzy Logic Scheduling scheme is proposed in this thesis. It is a hierarchical scheduling scheme. Advantages of Centralized Scheduling Scheme and Distributed Scheduling Scheme are integrated in this scheme. Bandwidth ranges of scheduled nodes is allocated periodically by the centralized scheduling part of Fuzzy Logic Scheduling. Bandwidth and priority of data are determined dynamically by the distributed scheduling part of Fuzzy Logic Scheduling. The disadvantage of traditional Centralized Scheduling Scheme is overcome in Fuzzy Logic Scheduling scheme, such as unacceptable performance degradation by fault of center scheduling node and network congestion in scheduling time. The flexibility and robust of Distributed Scheduling Scheme are embodied in Fuzzy Logic Scheduling scheme. Comparing with RM, MEF, EDF and MTS, the superiority performance of Fuzzy Logic Scheduling scheme is validated by experimental results.