Research And Simulation On Route Search Of Computer Interlocking System

As an important part of the train control system,interlocking system generally uses a computer-based interlocking technology to realize centralized control.System upgrade and increase of train's speed put forward higher requirements for safety and efficiency.Route control process of the computer interlocking system has a large effect on train transportation.Therefore,the thesis studies on the route search module of the route control process aimed to improve the safety and efficiency of train transportation.In the route search process,the traditional method is to search the interlock route table made manually in advance,but the search results still needs to do further comparison.This method has disadvantages of complicated process,low efficiency and high possibility of error.Besides,the original route table is no longer applicable and needs to be remade because of the station structure's change caused by the alteration of transportation demands and station structure's upgrade.Based on research on the route search process,the thesis first of all analyzes the route search process of the depth-first search algorithm.A problem has been found that there is a detour route existing when the depth-first search algorithm is applied.In order to avoid a detour route,constraints are set to improve the algorithm.A-Star algorithm is then discussed in detail to analyze its feasibility and advantages of route search application,and its heuristic function are determined through experiments.An interlocking simulation platform is built,and a directed graph model of the station is created by abstraction of the already designed station route data structure.With the use of Visual C++ programming platform,interface of the interlocking simulation platform is established.At the same time,two route search algorithms' internal logic realization is completed.Finally,route search experiments are conducted to testify the feasibility of those two route search algorithms and compare their performance.By analyzing the route search experiments' results,conclusions are made that these two algorithms can effectively find the correct route.In terms of time and number of the extend nodes,A-Star algorithm performances better and proves to be more effective in improving route conduction efficiency.