Velocity Optimization Of Urban Rail Automatic Train Based On State Space Model Evolutionary Algorithm

With the rapid development of cities and the rapid increase of population,urban traffic is facing more and more severe challenges.In this context,the automatic train operation(ATO)system emerges as the times require.It can replace the driver to realize the automation of train operation,thus reducing the driver's labor intensity,achieving energy-saving control,improving operation efficiency,ensuring positioning parking accuracy and improving ride comfort.Therefore,the optimization of ATO system has gradually become the key.Different train operation,corresponding speed protection target,punctual target,accurate parking target,comfort target and energy consumption target are also different,which requires the system to receive control center instructions,complete orderly operation and fixed-position parking between train stations,so as to realize automatic adjustment of train operation.In the current research of ATO system optimization methods,there are many problems,such as difficult optimization,incomplete information,difficult to find accurate models and optimal speed curve to describe train operation,and the results are different from the actual ones.In view of this,the thesis mainly completes the following work:Firstly,on the basis of in-depth analysis of the structure and function of ATO system,the control strategy and optimal operation principle in train operation are summarized,and a mathematical model describing the control performance objective of ATO system is established.However,in essence,the control of automatic train operation is a multi-objective optimization problem.The objectives of energy saving,safety,accurate parking and punctuality are often conflicting and need to be optimized at the same time.The ultimate goal of optimization needs to meet the requirements of energy saving,safety,accurate parking and punctuality,an improved state space model evolutionary algorithm is designed,which improves the selection operation.Individuals with high fitness are selected to enter the next generation.Individuals with poor fitness are re-assigned to the range of values and re-selected.Through this improvement,the search accuracy of the algorithm is improved.The validity of the improved method is verified by testing and analyzing two benchmark functions.By describing the parameters of train and line,the speed optimization simulation of train automatic operation is carried out on MATLAB platform with traditional state space model algorithm and improved state space model evolutionary algorithz,and the speed curve of running line is obtained.It is verified that the improved algorithm can meet the goals of punctuality,parking accuracy,energy saving and comfort at the same time when running in the range of train speed protection.Compared with the traditional evolutionary algorithm,the improved evolutionary algorithm can control the train better and improve the convergence speed and accuracy.