| The on-board equipment of CTCS-2(Chinese Train Control System Level 2)train control system belongs to the safety demanding system equipment,which must be strictly tested to ensure the safety of the system before it can be put into use.Test cases are the basis and guiding documents for the smooth progress of the test work and depend on the state,and the external performance of a test case is a switch from one state to another,and the input conditions of some test cases need to be achieved after executing other test cases.If the test cases are executed separately,it is necessary to continuously set the initial conditions for their execution,which increases the workload.Therefore,the test work at this stage will construct the test sequence based on the test cases with state matching according to certain rules.However,in order to achieve the principle of state matching between different test cases,there is usually a dynamic switching cost,which changes dynamically with the change of the execution state of the test process,such as the cost of controlling train acceleration/deceleration.In the existing test sequence optimization generation process,the dynamic switching cost between test cases is not considered,resulting in problems such as low test efficiency.Aiming at the above problems,considering the dynamic switching cost between test cases,this thesis studies the optimal generation of test sequence.The main work is as follows.1.In order to calculate the optimal dynamic switching cost between any two test cases,a test case dynamic switching cost estimation method is proposed.Firstly,the principle of train state transition brought by the execution of test cases is introduced.Secondly,according to the characteristics of train state transformation in the process of simulation and test,the constraint rules and cost calculation rules of train state transformation are defined.Thirdly,according to the state transition constraint rules and cost calculation rules,considering all possible state transitions,the network model of train state transition decision is constructed.Finally,a test case switching cost estimation algorithm is proposed to effectively estimate the minimum switching cost between test cases.The experimental results show that this method can fully cover the train state switching corresponding to various test cases,and quickly and effectively calculate the dynamic switching cost estimation results between any two test cases.2.Based on the above research point 1,a test sequence optimization generation method considering dynamic switching cost is proposed to construct the test sequence with the lowest total test cost.Firstly,based on the directed graph theory,the test cases to be tested are taken as nodes,and the dynamic switching cost between test cases is taken as the cost of interconnected edges between nodes.According to the characteristics of the dynamic change of execution state in the test process,the directed graph model of the test process is constructed.Secondly,aiming at minimizing the total switching cost between test cases,TSP-PSO(Traveling Salesman Problem-Particle Swarm Optimization)algorithm is designed to optimize the test sequence generation problem.Finally,some CTCS-2 on-board equipment to be tested cases are selected for optimal serial connection,which verifies the effectiveness of this method.3.Based on the above work,in order to improve the efficiency of simulation test,the test sequence optimization generation tool software of CTCS-2 on-board equipment is designed and implemented.The tool has the functions of test case management,test cost calculation and test sequence generation.It can effectively improve the generation efficiency and quality of test sequence in the simulation test process of CTCS-2 on-board equipment and reduce the test cost. |
PDF Full Text Request