Information Entropy-based Complexity Analysis Of Train Control System Architecture

The train control system is the core technology and equipment to assure the safety of train operation and improve the transportation efficiency. The scale of the current train control system is more and more large, complexity becomes increasingly higher, discrete calculation is tightly coupled with continuous physical process, running environment extends from closed and static environment to open, dynamic and persistently evolutive distributed network environment, there are a large number of uncertainties and heterogeneous subsystems. During the process of actual operation, because of the lack of clearer and more thorough understanding to train control system complexity, the safety-critical is not easy to be handled, and operation maintenance is under enormous pressure. In the design phase of architecture optimization for train control system, because of the lack of qualitative analysis and quantitative calculation for system complexity, the safety-critical is not easy to be controlled. Therefore, analyzing and measuring train control system complexity has profound theoretical significance and application value.In view of the lack of theoretical basis for designing and optimizing train control system architecture, and a lot of hardship in the safety-critical processes control during the system application, based on the information entropy theory, this thesis proposed a quantitative measurement method for the complexity of train control system architecture. The major research contents are as follows:(1) Expounded the research status of complexity science, the definition and analysis method of system complexity. On this basis, combining with the characteristics of the train control system, analyzed the complexity of the train control system architecture, and made the interpretation for the train control system architecture complexity from two aspects of structure and functional control logic.(2) Made a comparative analysis for the existing system complexity measurement, according to the characteristics of the train control system architecture itself, determined that the information entropy theory is used to research and analyze the complexity of train control system architecture, based on information entropy, established the measurement model for the complexity of structure and functional control logic.(3) Analyzed the CTCS-3of Wuhan-Guangzhou high-speed partial line, obtained the critical aspect that affected the system complexity and verified the effectiveness by using the measurement model for the complexity proposed.(4) From the perspective of reducing complexity of train control system architecture, three schemes for system structure optimization were proposed, based on the AADL modeling and verification tool, system architectures of different schemes were modeled and simulated for the same railway section, and verified their feasibilities from the respects of architecture, flow latency, resource budget and scheduling.The research of this thesis provides references for enhancing the operator’s capability of safety-critical process control, as well as for further optimization of the system.