Research For The Method Of Railway Block Signal Design Based On RailML

With the rapid development of Chinese railway industry,the demand for railway construction is increasing day by day.Interval signal design is a crucial part of railway construction,which includes two parts:the partition of blocks and the segmentation of track circuits.However,there are still many problems in the research of signal design in China and abroad.The basic train control data lacks a unified model,which causes low data interaction efficiency between different specialties and systems and causes high error rate.There is no multi-objective optimization research on block partition yet and the researches lack comprehensive optimization.For the track circuit segmentation problem,it is still designed mainly by hand.These problems have seriously affected the efficiency of railway signal design and hindered the development of railway industry.Aiming at the above problems,following works have been done:(1)Based on the basic train control data,a complete and accurate RailML model was built,which realized the unified specification of train control data.In addition,aiming at the problem that RailML parameters are not comprehensive enough,the model was extended appropriately to cover the necessary parameters of the follow-up work in the thesis,which lays the foundation for the subsequent signal design work;(2)According to the target and influence factors of block partition,the train tracking interval model was constructed.The thesis chose four-display automatic block as the research object.The objective functions are given respectively for "efficiency"and "economy".The block partition problem was researched from the perspective of multi-objective optimization,and was solved by Non-dominated Sorting Genetic Algorithm-II.(3)Based on the target and influence factors of track circuit segmentation,a mathematical model was constructed.According to the characteristics of track circuit segmentation,a two-step calculation method was proposed:the minimal number of segmentation points should be calculated first and the most average segmentation scheme be chosen next.The problem was solved by the Simulated Annealing Algorithm;(4)On the basis of the above research,a practical track was chosen to validate the method proposed in this thesis.First the RailML data model was constructed.And then the partition of blocks and the segmentation of track circuits was done using the above method.At last the signal data table was generated and the proposed method was validated.The innovations of the thesis are as follows:(1)The RailML model for track control data was constructed,which realized the uniform specification of track data;(2)The partition of blocks was accomplished from the perspective of multi-objective optimization,and the two objectives was optimized based on NSGA-?;(3)A two-step calculation method for track circuit segmentation was proposed,and the segmentation points were solved by the Simulated Annealing Algorithm.