Research On Modeling And Optimal Control For Following Operation Of High-speed Trains

High-speed railway is a national strategic infrastructure,an essential pillar of national economic development and an important public transport for people's daily life.It plays an important role in ensuring the effective implementation of national strategies,promoting the coordinated development of regional economies,and providing convenient and comfortable ways of life for people.While the high-speed railway in China has achieved great development significance and ushered in new development opportunities,some problems need to be solved urgently.With the dramatic increase in the frequency and operating speed of high-speed railways in China,the high-speed and high-density tracking operation of multiple trains in high-speed railway networks has become a new normal for high-speed railway operation management and the operation control of high-speed train(HST),especially prominent on railway lines.Existing train operation control methods are difficult to ensure that the tracking operation process of intelligent driving of HSTs meets multi-target operation requirements such as safety,efficiency,energy-saving and comfort.Therefore,it is one of the hot research issues in the field of intelligent driving control of HSTs to design a tracking characteristic modeling method and a multi-objective operation optimization control strategy adapted to high-speed and high-density tracking operation characteristics of HSTs.Based on the dynamic characteristics of HST tracking operation,the characteristics of operating scenarios,and the coupling characteristics of multi-train tracking operations,in this paper,effective and reliable models are established to describe HST's dynamic characteristics,the tracking run scene features and multi-train tracking characteristics.Besides,the robust speed tracking predictive control method and the multi-target coordinated optimization control strategy for multi-train tracking operation are proposed to ensure the safety,efficiency,energy-saving and comfort of HSTs' tracking operation process.The main work of this paper is summarized as follows:(1)The high-speed and high-density tracking process of HSTs has complex and variable disturbance factors and strong nonlinear dynamics.Traditional mechanical modeling methods are difficult to describe HST's dynamic characteristics effectively.Existing data-driven modeling methods have higher modeling accuracy,but it is difficult to guarantee modeling efficiency.It is necessary to provide a modeling method that can effectively and efficiently describe the dynamic characteristics of HST tracking operation.Echo state networks(ESNs)have significant advantages in effective and efficient modeling of nonlinear time-series,and are suitable for modeling typical nonlinear time series of HST operating data.However,ESNs have the problem of setting the random weight range singly and not adjusting based on data characteristics.Therefore,a unique assignment scheme for ESN's random weight matrix is proposed to achieve reliable and efficient modeling of the dynamic characteristics of HST tracking operation based on ESNs.(2)Accurate and stable speed tracking control is one of the key technologies for HST's intelligent driving.However,during the tracking operation of HSTs in an open environment,their operation control is constrained by many dynamic time-varying factors such as lines,the environment,their mechanical characteristics,and the running state of the front car,making precise and stable speed tracking control of HSTs challenging.The model predictive control method has advantages of efficiently handling dynamic multi-constrained optimization control problems and is widely used in train operation control research.And the reliability and efficiency of the prediction model are some the important factors that determine the performance of predictive control.Therefore,in this paper,the improved ESNs are used to establish the speed prediction model of HST,an adaptive adjustment strategy for the input temporal scale of ESNs model is designed to improve the model reliability further,so as to ensure the accurate and smooth speed tracking of the target speed curve during the tracking operation of HST.(3)In the tracking operation process of HST,line condition,signal system blocking system,electrified railway power supply and other scene specific factors are all important factors that affect the control force setting of train tracking operation.Therefore,effectively and reliably describing the main scene characteristics of HST's tracking operation process is an important guarantee for the validity and availability of the research results for HST's tracking operation optimization control.Also,it is one of the research hotspots in the field of train optimization control to give an efficient solution to HST operation optimization control problems based on scene feature models.Based on the analysis of the key features of the HST tracking operation scenario,a feature model of the HST tracking operation scenario including factors such as the safety distance constraint of the tracking operation,the profile of the line,and the electric phase separation inertia is established.Besides,based on this characteristic model,a multi-objective optimized control strategy for the tracking operation of HSTs is provided to achieve the safe,stable,energy-saving and comfortable tracking operation of a single HST.(4)The overall safe,efficient,energy-saving and comfortable operation of all trains in the railway section is an important objective for the optimal control of HSTs tracking operation.In the process of multi-train tracking operation,there is a strong coupling relationship between the running states of adjacent trains,and this relationship has a significant constraint on HST's control force setting.Therefore,accurately describing this kind of dynamic constraint relationship is the premise for conducting effective and reliable research on cooperative optimization control of multiple HST tracking operations.Existing research methods often only focus on the optimization control of single train tracking operation,or only consider the coordinated optimal control of multiple trains with the minimum tracking interval distance,but ignore the overall optimization goal of multi-train tracking operation control,as well as the situation that the scene characteristics of each HST may be different at the same time.This paper establishes a multi-train tracking characteristic model to accurately describe the constraint effect that the multi-train tracking operation state coupling relationship performed on HST's control force setting.Besides,a multi-train cooperative optimization control strategy with multiple goals of overall safety,efficiency,and energy-saving comfort in the line section is provided to achieve the safe,stable,efficient,and energy-saving multi-vehicle tracking operation of HSTs.