Research On Multi-Resolution Modeling Theory And Methods For Train Operation Control System

Operation process and control strategies generation in train operation control system involve crossing and integration of multiple disciplines. Because of complexity and distribution of train operation control system, it is complex and diverse in accuracy and detail level to create system simulation model. In this dissertation, based on multi-resolution modeling theory, CTCS-3train operation control system, which is currently widely used, was analyzed as an example. According to different simulation requirements, train operation control system was analyzed hierarchically and system models for various resolution levels were developed. Multi-resolution models for train operation control system could simulate system operation process under both normal situation and some special situations. Multi-resolution structural characteristics and multi-resolution modeling needs of CTCS-3train operation control system were analyzed. Three main research focuses of multi-resolution modeling theory, which are model description definition, modeling methods and multi-resolution model consistency maintenance, were thoroughly studied for modeling application in train operation control system. The idea of multi-resolution modelling was finally realized and verified on CTCS-3train operation control system simulation platform.The main innovation of this dissertation is as follows:(1) Multi-resolution model entities framework of train operation control system was proposed. Taking CTCS-3train operation control system as an example, train operation control process was analyzed hierarchically. Multi-resolution feature of simulation operation scenario and simulation system function was analyzed. Multi-resolution entities framework, including system macroscopic model, multi train tracking operation control model, single train operation control model and other simulation supporting models, as scenario simulation model and function simulation model, was proposed.(2) Model definition method of multi-resolution model space was proposed for train operation control system modeling. Concept and structure of multi-resolution model space were defined. Structures and functions of multi-resolution model group, interface model, resolution switching interface model and multi-resolution model space management model in multi-resolution model space were all set and described. In multi-resoluton model space, input and output of multi-resolution models could be managed thoroughly and generally. And also, by using interface model and space management model, existing simulation modules could be directly used to improve their resusability.(3) Multi-resolution entities aggregation and disaggregation method for train operation control system simulation was proposed. Based on the idea of the priority of intermediate layer, model entities of mid resolution level were created first and then model entities of lower and higher resolution levels were created. To realize resolution switch action dynamically and seamlessly during simulation, a new strategy, cross-resolution interaction coordination, was proposed. Based on train operation control system HLA simulation environment, three strategies, traditional resolution switching strategy, optimized pseudo-disaggregation strategy and new strategy, were realized and the applicabilities of these strategies were analyzed.(4) Consistency maintenance method of multi-resolution models output information for train operation control system was proposed. By using ontology mapping method, information atoms from different resolution model could be obtained in order to be compared mathematically. Output information from multi-resolution model family of train operation control system were analyzed and the mapping results, which were atoms for comparison, were obtained. In order to analyze the importance of the consistency of these atoms, consistency evaluation method for multi-resolution model output information was proposed based on FAHP method.Finally, multi-resolution modeling theory was verified on CTCS-3train operation control system simulation and testing platform in our lab. Train operation and control process could be simulated with multi-resolution representation on the constructed simulation and test platform. Macroscopic and microscopic events were all simulated on this platform successfully to verify the availability of the idea of multi-resolution modeling and simulation. A test plan was realized by real trains on real railway to obtain real train operation data. By simulating the same test plan on this simulation platform, comparison results between simulation data and real train operation data proved the rationality and credibility of multi-resolution modeling and simulation of train operation control system.