Research On Test Sequence Generation And Optimization Method Of C3+ATO System

As the rapid development of national economy,high-speed railway is widely favored because of its high comfort,high travel efficiency and high safety factor.Meanwhile,according to the national planning outline for a comprehensive three-dimensional transportation network,it is expected to build a total of about seventy thousand kilometers high-speed railway network around 2035.Therefore,China’s railway is expanding towards the intelligent goal of high-speed and high-quality.On the basis of CTCS-3(Chinese Train Control System),our country has independently developed the CTCS-3+ATO(abbreviated as C3+ATO)system of high-speed railway train control system with ATO(Automatic Train Operation)function which has the feature of high safety requirements,high speed and high automation,which will be an important application development direction of intelligent high-speed railway.Therefore,it is of great significance to verify and test the function of C3+ATO system to ensure the safe control operation.As the guidance document of train control system function test,the quality of test sequence is related to the final test results.However,the current test sequence of train control system mainly depends on the manual preparation of testers,which has not only low efficiency,but also low availability.Therefore,the research on test sequence generation and optimization of C3+ATO system is closely related to ensuring system functional safety and improving test efficiency.The typical functional scenes of C3+ATO system are taken as the subject in this thesis.The test method based on formal model is used to establish the scene models,simulate and verify the function attributes of the model.The test sequences are generated by parsing the model file and optimized by using Firefly Algorithm(FA)and its improved algorithm.The main research contents include:(1)The scene function of C3+ATO system is analyzed.The scene function model of C3+ATO system automatic departure at the departure station and on-board equipment mode transition are established based on timed automata theory.The model is simulated and verified by generating the message sequence diagrams of the corresponding functional description and verifying the functional attributes of the model.By comparing with the relevant technical specifications of the C3+ATO system,the established models meet the functional attributes of the corresponding scene.(2)The Tiny XML parsing library is used to parse the generated model files to get the test case sets.The test case generation principles are set to concatenate the test cases that meet the test requirements to generate test sequences.In order to improve the efficiency of testing,the VS 2015 is used to design a test case set and test sequence automatic generation tool.When the main states of the test sequence are set and the coverage standard is selected,the test sequences meeting the test requirements can be automatically generated.(3)Taking the repetition rate and algorithm convergence rate of the test sequence which are related to the improvement of test efficiency as the optimization criteria,the firefly algorithm is proposed to optimize the generated test sequences.The firefly algorithm can reduce the repetition rate of the test sequence of the automatic departure scene at the departure station by 15.4%.In addition,the firefly algorithm is improved by improving the step size factor.And the simulation results demonstrate that the improved firefly algorithm can efficiently improve the convergence rate.