Font Size: a A A

Research And Application Of Performance Analysis, System Design And Network Interconnection In Control Network System

Posted on:2005-10-21Degree:DoctorType:Dissertation
Country:ChinaCandidate:K PengFull Text:PDF
GTID:1118360125958070Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
As the applications of large scale, complex and distributed control continue to grow in both daily life and social production, computer and network communication technologies are widely used in the domain of automatic control. Merging of control, computer and network technologies promotes the birth and development of control network system (CNS), which consists of a brand new control pattern. Control network technology has caused profound changes in control area and CNS is due to be the mainstream of future automatic control systems. Nevertheless, at this moment, CNS has not yet been a research area with full-fledged architecture, efficient technical methods and clear extensions. There are still a lot of theoretical and practical problems to solve, especially the problem that its theory is dramatically inconsistent with practical applications. On this background, the dissertation presents some research work and discussion about the key problems of performance analysis, system design and network interconnection in CNS. It also contains analysis and synthesis of technologies and applications related to this work. The author's own viewpoints, analytic methods and solutions are detailed. The whole column is divided into six chapters.The first chapter introduces the definition of control network and CNS, discussing their characteristics and advantages. On the basis of analyzing the development of networked control systems, it summarizes three important development phrases of CNS and their own technical characteristics. Then, it studies current situation of control network development, pointing out that interconnection and integration of different kinds of CNS is the key characteristic of the phrase undergoing. Finally, the chapter explains, in detail, the research status of CNS in both theory and application.The second chapter discusses the QoS of protocols used in CNS. First, it utilizes diagrams to compare and analyze the static QoS of three CSMA control network protocols, which are Ethernet, CAN and LON, proposing that static QoS analysis can be used to choose the type of control network and it is valuable to early stage of CNS design. Secondly, by studying the composition of network transmission delay, it concludes that network delay and communication reliability depend mainly on data link layer, especially MAC layer of the protocols. Then this chapter goes a step further to use analytic methods and simulations to get the dynamic QoS of CSMA, Token Bus and Master-Slave Polling protocols. It also suggests a simulation method to study dynamic QoS, with the use of finite automata and Rational Rose RealTime modeling software. At last, the chapter proposes two modifications of CSMA protocol: priority rotated CSMA protocol and switched CSMA/CD -Master/Slave Polling hybrid protocol. With theoretical analysis and simulations, it proves their advantages in real-time and reliability.The third chapter focuses on three sub issues in quality of control performance (QoCP) analysis in CNS: 1. Using nonlinear programming, it solves the problem of how to optimize sampling frequencies in multi-loop CNS, and further concludes algorithm for practical engineering use. 2. As to different types of network transmission delay and packet loss, it obtains the stability conditions for CNS. 3. By setting up receive buffers, it makes therandom network delay certain, then completes the design of a two-degree-of-freedom internal model controller (IMC) with optimized parameters based on chaos optimization algorithm in CNS.The fourth chapter discusses methodologies in designing CNS: 1. On the design of underlying control network nodes, the dissertation proposes to use hardware and software co-design method based on the structure of multi-processor and function modules. This method greatly improves the capability, extensibility and reliability of a single node. The method has been applied to develop a multi-function LonWorks node with CPLD and multi-microprocessor architecture, which has been successfully used in practical projects. 2. By...
Keywords/Search Tags:Control Network System(CNS), Quality of Service(QoS), Quality of Control Performance (QoCP), formalized modeling, system design method, network interconnection
PDF Full Text Request
Related items