The Algorithm Of Test Sequence Optimization Based On The Improved Ant Colony Algorithm

In recent years, with the rapid development of China railway operation control system, the improvement of stability and safety have become the focus of the development of CTCS-3train control system. In order to improve the reliability of CTCS-3train control system, ensure its smooth running and avoide unnecessary loss, a large number of tests must be completed before its formal operation. However, the tests will consume a large amount of manpower and material resources, and the system test cycle is relatively long, what’s more, some loopholes and defects will exist. Therefore, the model-based automated test method has become an attractive approach for ensuring the correctness of CTCS-3function and improving the efficiency of the test.In this paper, the improved ant colony algorithm (Mouse Maze Algorithm-Ant Colony Algorithm, M-ACA) has been proposed, which combines mouse maze algorithm and ant colony algorithm. It has been applied to the study of the automatic method of test cases and test sequences based on the CPN model. Four scenes,which include RBC handover, auto-passing phase, registraton and start, canceled, are selected as examples to make a scientific validation and detailed exposition combined with the RBC test platform. The main work of the paper is as follows:Firstly, a brief overview of research status, background and significance of automatic test is made and several automated testing methods based on the model are enumerated.Secondly, the related concepts are introduced, including the research of test method, the analysis of the traditional algorithm, the relevant concept of CPN and CPN Tools.Thirdly, the basic principle of M-ACA has been proposed.Then the process of generation and optimization of test case and test sequence is elaborated according to the flow chart. In addition, the process of building the CPN model is explained in detail. The generated results and the process of generating optimal sequences are introduced through state space reachability graph.Fourthly, according to the improved algorithm, the test sequence generation software has been built using C++programming language which is made up of demanding module, testing module, operation module, result module. Comparing the optimization results obtained by running the software with the test sequence generation results of using the sequence priority selected algorithm (SPS) and the traditional depth first search algorithm (DFS), the superiority of the proposed M-ACA algorithm is ultimately proved. In order to meet the needs of the actual operation, the fault and parallel test methods are added, and the test sequence generation tool based on fault and parallel test sequence generation tool are made. The results of corresponding generated test sequences are compared and analyzed.Finally, the algorithm and the program would be validated by connecting to the RBC platform. The purpose of realizing the automatic generation and optimization of test sequence can be accomplished. The algorithm not only removes the redundant parts of test sequence, but also reduces the multiplicity. What’s more, it covers the basic needs of the relevant contents of the CTCS-3system specification.