Study On Operation Adjustment Of High-speed Trains Based On Genetic Simulated Annealing Algorithm

The high-speed rail is popular with the public for its fast,comfortable and convenient features.However,with the increase in travel speed,people have put forward higher requirements for stability and punctuality.Usually,trains operate strictly according to the time specified in the operation diagram.However,during the operation of the train,it will inevitably be affected by various factors and and deviate from the operation diagram,which will cause delay.The corresponding adjustment should be made in time to make the train return to the punctual operation as soon as possible,otherwise it may cause a large range of delay,causing serious safety hazards and economic losses.The traditional train operation adjustment in China is manually completed by dispatchers.This method has high work intensity and high requirements for dispatcher's professional ability.Train operation adjustment is a large-scale combinatorial optimization problem with many parameters and complex solutions.At the same time,trains on the high-speed railway are intensive,fast,and have high punctuality requirements,thus traditional methods are difficult to fulfil real-time requirements.Therefore,research on fast,intelligent and automated train operation adjustment is necessary.With the development of new technologies,intelligent algorithms have provided new ideas for the study of this problem in recent years.In this thesis,the high-speed train operation adjustment is taken as the research object,and the following research is carried out on the basis of previous studies.First,train operation diagram is described in a mathematical manner and is used as the basis of modeling.The train operation adjustment model is established by describing constraints in the train operation adjustment process mathematically and taking the total weighted arrival delay time of the section as the objective function in consideration of the train level and adjustment weight.Then,based on the characteristics of train operation adjustment,the genetic algorithm is selected as the basis and combined with the simulated annealing algorithm to obtain the genetic simulated annealing algorithm.This algorithm has comprehensive search capabilities and combines the advantages of both.Combining with the specific problem,this thesis designs and implements the genetic simulated annealing algorithm,and designs corresponding solutions for the violation of the constraint conditions in the solution process.In order to verify the effectiveness of the genetic simulated annealing algorithm and analyze its performance,the actual train operation time information of Beijing-Shanghai high-speed railway is taken as an example to compare genetic simulated annealing algorithm's adjustment results with genetic algorithm's adjustment results as well as interval acceleration operation method's adjustment results.Finally,based on the virtual simulation experiment platform of Beijing-Shanghai high-speed railway train control system,the train operation adjustment simulation system is designed and programmed with C#programming language and MySQL database.Using the system,the genetic simulated annealing algorithm is selected to adjust the train operation for two different delay situations,to verify the effectiveness of genetic simulated annealing algorithm for solving the train operation adjustment problems at different delay situations,and to comprehensively verify the effectiveness of the core train operation adjustment function of the system.The results of this research indicates that genetic simulated annealing algorithm can effectively solve the train operation adjustment problem under different delay situations and alleviate the delay condition of the train.Compared with the genetic algorithm,the algorithm improves the local search ability and enhances the directionality of the algorithm search by adding the simulated annealing step.So that the solution is more accurate and the convergence speed is faster,which is manifested in the shorter adjustment time and smaller total weighted arrival time value.The train operation adjustment simulation system developed in this thsis fulfils a complete set of functions designed,and its core train operation adjustment function has satisfying adjustment effects for different delay situations.