The Research And Simulation Of Automatic Train Operation Algorithm

As a key technology, the train control system plays a significant role in the railway construction. The exsiting train control system is only designed to control the speeding trains and ensure the safety. However, the construction of high-speed railway is advancing at an amazing pace and there is growing pressure on energy saving and stop accuracy. The exsiting control system will not meet the demands of the railway development. Thus, the new optimal train control system should not only ensure the safety, but also improve the efficiency intelligently. In this situation, there are a great deal of research about the Automatic Train Operation(ATO), which can replace human operators and improve the driving strategy, energy conservation,comfort level, the gap of stop accuracy and other aspects. In this research, it learned the ATO system from the urban mass transportation, then it worked out a new ATO algorithm to train mainlines and made a simulation system to verify it.In this dissertation, firstly it researched the background of ATO. Next, according to the feature of train mainlines; it analyzed the functional requirements, and made a proposal to add the ATO subsystem into the exsiting train control system. After that, it worked out the ATO algorithm, which includes the identification of performance indices, the summarization of driving strategies, the application of fuzzy control rules, the fuzzy set of performance indices, and the application of Analytic Hierachy Process to solve the conflicts among the performance indices. Then, it made a set of ATO control rules. With the rules, it could make the decision of the best control command. The next step is to design the ATO simulation system in the object-oriented programming language; which is based on the Chinese Train Control System level 3 and the data of CRH series trains and the data of Hefei-Nanjing high speed line. At last, it compared the simulation function of manual driving and automatic driving, and then it verified that the ATO algorithm is more efficient in improving the driving indices.