| With the increase in travel demand,China’s high-speed train technology has developed rapidly,and the eight vertical and eight horizontal high-speed railway network is also gradually being built.High-speed trains have become an important part of modern transportation because of their high speed,high operational efficiency,low carbon and good riding comfort.Therefore,it is crucial to design a safe and reliable speed control system for high-speed trains so that they can run smoothly with high precision according to the desired speed and displacement curves.The main research and results of this paper include the following:Firstly,a multi-particle high-speed train dynamics model with time-varying coefficients is designed for the problems of uncertainty disturbances generated by the operating environment,the mutual coupling forces between carriages and the uncertainty of the internal coefficients of the train during the operation of high-speed trains.The working principle of high-speed trains’ automatic train operation system is introduced in detail,and the establishment process of the longitudinal dynamic model is analyzed to provide a model basis for verifying the designed high-speed trains speed control scheme.Secondly,the linear active disturbance rejection control(LADRC)is adopted as the speed control scheme of high-speed trains.This control scheme does not rely on accurate mathematical models.It uses linear extended state observer(LESO)to estimate uncertain external disturbances and internal unmodeled disturbances in the operation of high-speed trains.The disturbances are compensated by the proportional derivative(PD)controller to improve the anti-disturbance performance and control accuracy.On this basis,the linear differential tracker is introduced to filter and smooth the input signal to solve the problem of noise and overshoot caused by the input signal.Thirdly,the adaptive mutation particle swarm optimization(AMPSO)is used to adjust the parameters of multiple LADRCs in high-speed trains according to the speed error of each carriage.This greatly simplifies the process of parameter setting and ensures that the controller of each LADRC has good control effect.This paper introduces the optimization process of the traditional particle swarm optimization,particle swarm optimization with improved weight and AMPSO,and verifies the advantages of AMPSO as LADRC parameter setting tool through simulation experiments.Fourthly,considering the problems that traditional LADRC cannot adjust the parameters adaptively according to the system environment,and the observation error caused by the bandwidth limitation of LESO,adaptive control(APC)is introduced and adaptive linear active disturbance rejection control(ALADRC)is proposed.The adaptive laws are designed for LADRC parameters and LESO,so that the controller parameters can be adjusted adaptively according to the tracking error of the high-speed trains,and the observation error can be compensated at the same time.The introduction of APC enables the system to be validated by the Lyapunov stability theory,and the superiority of this control scheme compared with LADRC is verified by theoretical analysis and simulation.Fifthly,in order to further improve the stability of high-speed trains operation,considering the uncertainty of the internal coefficients of the model caused by the highspeed trains self factors and operating environment,a model-based ALADRC composite control scheme is proposed based on the high-speed trains model.The adaptive laws are designed for the time-varying coefficients of high-speed trains model.The real-time compensation of the adaptive laws can effectively offset the adverse effects of timevarying coefficients on high-speed trains operation.The advantages of the composite control scheme compared with ALADRC is verified by detailed theoretical analysis and simulation experiments.The stability of the system is verified by the Lyapunov stability theory. |