Research On Automatic Analysis Method For Simulation Integration Test Results Of Train Control System

The train control system is the key basic equipment to ensure the safe and efficient operation of the railway,and it is also an important part of high-speed railway construction.As a signaling control system with extremely high safety and reliability requirements,it must be tested completely in various scenarios.The simulation integration test of the train control system is a key link to verify whether the system meets the standard specifications and engineering design requirements.The test scope covers the system function test,system interface test and engineering data test.Through the analysis of the simulation integration test results,the evaluation of the train control subsystem or the entire system can be achieved.At this stage,the analysis of test results is mainly for the testers to manually judge the test items based on design requirements,test specifications,and expert experience.This method is inefficient and requires high technical capabilities for testers,which can easily lead to misjudgment of test results.In order to solve these problems,this thesis studies the automatic analysis method of the simulation integrated test results for ground equipment of the train control system.The main research contents are as follows:(1)In order to meet the needs of automated testing better,using virtualization technology to optimize the design of the existing train control system simulation integrated test platform from the aspects of functional architecture and hardware environment.(2)The overall scheme of automatic analysis of simulation integration test results is studied.Through the acquisition,classification and dimensionality reduction of the line data,the status data for ground equipment of the train control system and the on-board record data,and use it as the data source of the automatic analysis method of the test results;according to the differences in the types of items examined during the test,the application of rule reasoning based on expert system knowledge base and KNN algorithm in automatic analysis methods are studied respectively.(3)The function realization of the test result automatic analysis tool is designed and structurally divide the tool into three levels: data processing layer,logic operation layer and interactive interface layer.In the data processing layer,the test data acquisition method,feature extraction method and storage structure are designed.In the logic operation layer,the automatic analysis algorithm based on rule-based reasoning is used to design the test result analysis process of several typical test scenarios in detail;the automatic analysis algorithm based on K-nearest neighbor is designed for the permitted speed of the vehicle.In addition,the main functional requirements of the interactive interface of the test results automatic analysis tool are analyzed and designed.(4)The test results automatic analysis tool is realized from the aspects of feature extraction and database access of test data,and the virtual simulation integrated test platform is built using vSphere technology to complete the deployment of the test results automatic analysis tool.In the train control system simulation integrated test environment of an actual line,several typical test scenarios are used to verify the correctness and usability of the automatic analysis method of the test results proposed in this thesis.This thesis contains 54 pictures,17 tables,and 55 references.