Research On The Expert System For Generation Of Test Sequence For CTCS-3Train Control System

CTCS-3is an important technical equipment to meet the high safety demand of railway transportation, which has been successfully put into service in several300km/h high speed railways, such as Wuhan-Guangzhou, Zhengzhou-Xi’an, Shanghai-Hangzhou, Shanghai-Nanjing and Beijing-Shanghai high speed railways. A series of testing including laboratory simulation testing, field testing, integrated testing and commissioning and interoperability testing must be carried out with the aim to verify whether the system complies with the technical specifications and whether the train equipped with the onboard equipment of one supplier can run on the lines equipped with RBC from other suppliers. Therefore it is necessary to prepare various kinds of test sequences according to "CTCS-3test cases" issued by the Ministry of Railways. At present, such sequences are worked out manually by experienced testers with complex procedure and low efficiency. In view of this, a computer aided test sequence generation tool has been developed by Beijing Jiaotong University, which can provide a graphic editing environment, and can automatically generate the WORD format test sequences for field testing and the XML test scipts used in laboratory simulation, thus greatly reduce the effort for generating test sequences.This thesis proposed a test sequence genration expert system on the basis of the test sequence computer aided generation tool in order to improve the intelligence of the tool Firstly, on the basis of summarizing the advantages and disadvantages of the test sequence generation tool, the feasibility and advantage of using expert system to aid the generation of test sequence is put forward. Secondly, through discussion with experts and with the help of technical staff, the expert knowledge for test sequences generation is collated, and knowledge base is established. At the same time, the knowledge is represented by means of logical operation, semantic network and so on. Then simple reasoning is carried-out with the method of production rules, knowledge tree, thus basically complete the function of test sequence generation expert system. Finally, based on the production rules, and backward reasoning method to carry out test sequence with specific case.