Automatic Train Operation System Research Articles

Data-driven models for train control dynamics in high-speed railways: LAG-LSTM for train trajectory prediction

The construction of an accurate train control model (TCM) is crucial to the design of automatic train operation and real-time traffic management systems in high-speed railways. Traditional physical-driven models usually fail to reflect the “true” dynamics of high-speed trains (HSTs) because of the strong nonlinearity and uncertainty due to air resistance, frequently switching working conditions, and variations in external influencing factors such as weather and temperature. Although some data-driven deep learning models have recently been proposed for environmental adaptation, they are all “black-box” models, which cannot explain how the input of the models affects the HST output. To overcome these issues, this study constructs a novel long short-term memory with lagged information (LAG-LSTM) model by combining the physical-driven HST model and an “interpretable” deep learning model. Specifically, our LAG-LSTM model contains three modules: a time-delay variable module to model the transform delay of control variables, state variable enhancement module to extract the key features among high-dimensional input data, and Pre-LSTM module to predict the future train trajectory with given control variables. We collected field data from the Beijing-Shanghai high-speed railway and developed a data filter method and a normalization procedure to overcome the positioning errors of HSTs and construct a standard data set. Finally, we tested the effectiveness of our LAG-LSTM by comparing it with six deep learning structures, including fully connected neural network, recursive neural network, standard LSTM, and LSTM with convolutional layers. The results show that LAG-LSTM can accurately predict the trajectories of HSTs and outperforms other deep learning models. Regarding prediction accuracy, LAG-LSTM improved the performance of the traditional LSTM by 13.5% to 23.3%.

Research on Generating Method of On-board Equipment Test Sequence in ATO System of High Speed Railway

With the development of railway technology, the Train control system of Beijing-Zhangjiakou high-speed railway, which was opened in 2019, applies CTCS-3+ATO (Automatic Train Operation) technology to realize automatic driving at 350km/h, including automatic train opening, automatic and precise parking, and the linkage between train doors and platform doors. To ensure the safety and reliability of train control system, it is necessary to test the system. In this paper, the test cases used in the testing work are concatenated into test sub-sequences according to certain principles, and the concatenation problems of test sub-sequences are converted into TSP (Travelling Salesman Problem) for solving. PSO (Particle Swarm Optimization) algorithm was used to obtain the test sequence of ATO system in high-speed railway, which provided guidance for the test work.

Development of energy evaluation system during manual operation using train simulator

AbstractThe electric railway is an environment‐friendly transportation system. An energy‐saving operation with a low cost and high energy‐saving efficiency is often considered to further decrease energy consumption. An Automatic Train Operation (ATO) system has the advantage of energy‐saving because a train can be controlled by an on‐board computer. However, many railway lines are still being operated by drivers and controlled via manual operation in addition, energy managements often become difficult for drivers and operators. Therefore, a study of energy‐saving manual operation is desired. Energy verification of a manual operation is difficult because an adequate experiment has several constraints and trial periods are limited. Moreover, it requires a considerable amount of time and effort. In this study, an energy evaluation system is developed to calculate the energy consumed by a train when a driver manually operates it through adding an advanced calculation device to a conventional railway simulator. Consequently, the evaluation system contributes to carry out simple experiments with a short duration.

Automatic control of computer application data processing system based on artificial intelligence

Abstract To shorten the travel time and improve comfort, the automatic train driving system is considered to replace manual driving. In this article, an automatic control method of computer application data-processing system based on artificial intelligence is proposed. An automatic train operation (ATO) introduced the structure and function of an autopilot system (train), optimized the train running on the target curve, introduced the basic principle of fuzzy generalized predictive control (PC) algorithm, and combined with the characteristics of ATO system design the speed controller based on optimization algorithm, the target curve to make use of the designed controller to track and simulation validation. The experimental outcomes demonstrate that when the train runs to 90 s, the displacement difference reaches about 40 m, which proves that the fuzzy PC has better displacement tracking, punctual arrival, and higher stopping accuracy.

ОПРЕДЕЛЕНИЕ ПАРАМЕТРОВ ПОЕЗДА В ПУТИ СЛЕДОВАНИЯ ПО ДАННЫМ БОРТОВЫХ ИЗМЕРИТЕЛЬНЫХ СИСТЕМ ГРУЗОВЫХ ЭЛЕКТРОВОЗОВ

The paper suggests an approach to determination of actual train operation parameters according to data from on-board measurement systems in order to increase efficiency of application of automatic electric motive power operation systems. The authors have analysed scientific works in the sphere of improvement of automatic electric locomotive operation. The paper also presents results of comparison of grade computations in KORTES and real data from on-board measurement systems on 2ES6 electric locomotives that operate on a section of the West-Siberian Railway. As a result, the authors have substantiated a possibility to apply data gained from on-board measurement systems in real-time mode for determination of such train parameters as tractive effort torque of traction motors, electric locomotive traction effort, external forces of train tractive resistance, reduced gross load hauled and etc. The approach to determination of train parameters according to data of current measurements allows increasing peculiarity of transmission of control actions and efficiency of an automatic train operation system.

On the safety assessment of an automatic train operation system

The paper examines the automatic train operation system as part of the locomotive control and protection system, the remote supervision centre’s means for control of onboard and trackside machine vision facilities. The focus is on the dependence of the system’s safety and dependability on the dependability characteristics of its components and adverse weather effects. The criteria of a system’s wrong-side and right-side failures were defined, the graph models were constructed of the safety and dependability states of an automatic train operation system. The Markovian graph method of calculating the safety and dependability of complex systems was substantiated. That allowed defining such key safety indicators of an automatic train operation system as the mean time to wrong-side failure, probability of wrong-side failure, wrong-side failure rate. The study established that the safety of an automatic train operation system primarily depends on the dependability of machine vision facilities. The growth of the system’s wrong-side failure rate is limited to half the failure rate of machine vision facilities. It was also established that the dependability of an automatic train operation system is defined by the failure rate of a locomotive control and protection system and the failure rate of machine vision facilities. The conducted analysis allows concluding that in order to achieve an acceptable level of safety of an automatic train operation system, efforts should focus on machine vision redundancy, ensuring the SIL4 functional safety of on-board and trackside machine vision facilities, as well as regular comparison of the outputs of on-board and trackside machine vision facilities, redundant output comparison, integration of the outputs in motion. Additionally, adverse weather effects are to be countered by improving the efficiency of machine learning of the machine vision software.

Organizational and economic mechanism of the automatic underground train operation system

This work deals with the provision of underground, as one of the main modes of public transport, with automated control systems, in particular the provision of automated train management. It defines that the underground is one of the most progressive modes of transport, which is related to its environmental safety, comfort, and speed of movement, as this mode of transport does not intersect with other transport and pedestrian routes, which excludes traffic blocks. Underground is an important element of urbanized spaces, for it serves large passenger traffic. The article presents an algorithm for the creation of an automated control system for the underground train system. Based on this algorithm, a model of building an organizational and economic mechanism for automation of control systems has been developed. The diagram of organizational and technical implementation of the automated system of underground train management is presented. An economic analysis of the effectiveness of the application of the automated control system of underground trains was carried out. The recoupment on the acquisition and installation of automation systems is very high. The application of the automated system will increase the capacity of the underground by optimizing the traffic schedules, which will contribute to increasing its profitability. It is also useful to determine the reduction of electricity consumption due to the change in the dynamics of acceleration and braking systems, which is important, as the underground is a powerful enterprise. In the future, a promising direction is to streamline the coordination of train schedules and other modes of land public transport.

Notch-based speed trajectory optimisation for high-speed railway automatic train operation

Automatic train operation (ATO) systems are fast becoming one of the key components of the intelligent high-speed railway (HSR). Designing an effective optimal speed trajectory for ATO is critical to guide the high-speed train (HST) to operate with high service quality in a more energy-efficient way. In many advanced HSR systems, the traction/braking systems would provide multiple notches to satisfy the traction/braking demands. This paper modelled the applied force as a controlled variable based on the selection of notch to realise a notch-based train speed trajectory optimisation model to be solved by mixed integer linear programming (MILP). A notch selection model with flexible vertical relaxation was proposed to allow the traction/braking efforts to change dynamically along with the selected notch by introducing a series of binary variables. Two case studies were proposed in this paper where Case study 1 was conducted to investigate the impact of the dynamic notch selection on train operations, and the optimal result indicates that the applied force can be flexibly adjusted corresponding to different notches following a similar operation sequence determined by optimal train control theory. Moreover, in addition to the maximum traction/braking notches and coasting, medium notches with appropriate vertical relaxation would be applied in accordance with the specific traction/braking demands to make the model feasible. In Case study 2, a comprehensive numerical example with the parameters of CRH380AL HST demonstrates the robustness of the model to deal with the varying speed limit and gradient in a real-world scenario. The notch-based model is able to obtain a more realistic optimal strategy containing dynamic notch selection and speed trajectory with an increase (1.622%) in energy consumption by comparing the results of the proposed model and the non-notch model.

Optimization of Urban Rail Automatic Train Operation System Based on RBF Neural Network Adaptive Terminal Sliding Mode Fault Tolerant Control

Aiming at the problem of the large tracking error of the desired curve for the automatic train operation (ATO) control strategy, an ATO control algorithm based on RBF neural network adaptive terminal sliding mode fault-tolerant control (ATSM-FTC-RBFNN) is proposed to realize the accurate tracking control of train operation curve. On the one hand, considering the state delay of trains in operation, a nonlinear dynamic model is established based on the mechanism of motion mechanics. Then, the terminal sliding mode control principle is used to design the ATO control algorithm, and the adaptive mechanism is introduced to enhance the adaptability of the system. On the other hand, RBFNN is used to adaptively approximate and compensate the additional resistance disturbance to the model so that ATO control with larger disturbance can be realized with smaller switching gain, and the tracking performance and anti-interference ability of the system can be enhanced. Finally, considering the actuator failure and the control input limitation, the fault-tolerant mechanism is introduced to further enhance the fault-tolerant performance of the system. The simulation results show that the control can compensate and process the nonlinear effects of control input saturation, delay, and actuator faults synchronously under the condition of uncertain parameters, external disturbances of the system model and can achieve a small error tracking the desired curve.

A New Estimation Method of the PID Controller In Automatic Train Operation Systems

PID controller plays an important role in Automatic Train Operation systems. Many researches have been done on the PID parameters estimation. However, the traditional models simplified the characteristics of the trains in different speeds and operation modes. It was difficult to overcome the problems with different kinds of algorithms based on the traditional models, especially the accuracy requirement of the stopping distance. In this paper, a new model of the PID controller was proposed based on the practical train operation stages. And the different response time delay was considered besides of the transmission time delay. With this new model, an Improved Fruit Fly Optimization Algorithm was applied. The tracking performance were greatly improved both on running distance and real time. The simulation results showed a better performance than the traditional methods.

Research on Automatic Braking and Traction Control of High-speed Train Based on Neural Network

With the rapid development of economy and technology, people’s quality of life has been continuously improved, which also makes people put forward higher requirements for the quality of transportation. As far as railway transportation is concerned, the running speed and route of trains in the past gradually failed to meet people’s travel needs. Automatic train operation system (ATO), under the protection of automatic train protection system, realizes automatic train driving according to the instructions of automatic train supervision system, and can automatically control the starting, traction, cruising, idling and braking of trains. Traction drive system is the core part of high-speed train, and traction drive converter technology is one of the key technologies of high-speed train. In this paper, according to the different working mechanism of traction and braking, the dynamic model of train under traction and resistance braking is established by combining neural network algorithm, and the corresponding controller is designed by using nonlinear backstepping design method to realize automatic tracking of train speed and position.

Integrated train dwell time regulation and train speed profile generation for automatic train operations on high-density metro lines: A distributed optimal control method

The wide-spread application of automatic train operation (ATO) system on metro lines allows short service headways, high-density operations and high operation efficiency. This paper addresses real-time train control for ATO when faced with disturbances or disruptions in its operations. More specifically, the paper focuses on the design of integrated train dwell time regulation and speed profile generation in real-time and in response to dynamic changes in the operation environment. A nonlinear optimal control model is formulated in a rolling horizon scheme that incorporates three key operating elements: train timetable, passenger load and train speed profile. The objective is to simultaneously improve headway regularity and reduce the total energy consumptions. To satisfy the real-time control requirement for ATO system, a decomposition method based on the alternating direction method of multipliers (ADMM) is designed to divide the original optimization problem into many sub-problems, one for each train, which can then be computed in a distributed manner. Moreover, to address the non-convexity issue, a relax-round-polish process is developed to deal with the formulated nonlinear optimal control problem with convex objective over non-convex constraints in order to find the approximate solutions quickly for the embedded applications. The combined result is an ADMM-based heuristic algorithm. The effectiveness of the proposed model and solution algorithm is demonstrated using real-world data from the Changping Line of Beijing Metro. The results show that the proposed distributed and embedded optimization algorithm is able to significantly enhance the robustness and reliability of real-time train control in automated high-density metro lines.

Multiobjective Optimization on the Operation Speed Profile Design of an Urban Railway Train With a Hybrid Running Strategy

In urban railway systems, a predefined speed profile helps train drivers and automatic train operation systems realize eco-driving without ignoring the requirements for punctuality and comfort. This study proposes a multiobjective optimization method for speed profiles of urban railway systems and provides a Pareto front in three dimensions with a hybrid running strategy. First, three popular running strategies are selected and combined. Second, a model of train behavior is constructed, and energy efficiency, running time, and comfort are used to evaluate the speed profiles. Third, a hybrid strategy multiobjective particle swarm optimization algorithm is proposed and utilized with train performance simulation to solve this problem. Finally, two running sections of Beijing subway line 8 are applied in a case study. The results verify the effectiveness of the proposed approach and imply that a single strategy should not be abandoned when comfort is a consideration.

Analysis of ATO System Operation Scenarios Based on UPPAAL and the Operational Design Domain

With the gradual maturity of the automatic train operation (ATO) system in subways, its application scope has also expanded to the high-speed railway field. Considering that the ATO system is still in the early stages of operation, it will take time to fully mature, and definite specifications of the requirements for system operation have not yet been formed. This paper presents the operational design domain (ODD) of the high-speed railway ATO system and proposes a scenario analysis method based on the operational design domain to obtain the input conditions of the system requirements. The article models and verifies the scenario of the linkage control of the door and platform door based on the UPPAAL tools and extracts the input and expected output of the system requirements of the vehicle ATO system. Combined with the input conditions of the system requirements, the system requirements of the vehicle ATO in this scenario are finally obtained, which provides a reference for future functional specification generation and test case generation.

Data-Driven Koopman Model Predictive Control for Optimal Operation of High-Speed Trains

Automatic train operation systems of high-speed trains are critical to guarantee operational safety, comfort, and parking accuracy. However, implementing optimal automatic operation control is challenging due to the train's uncertain dynamics and actuator saturation. To address this issue, this paper develops a data-driven Koopman model based predictive control method for automatic train operation systems. The proposed control scheme is designed within a data-driven framework. First, using operational data of trains and the Koopman operator, an explicit linear Koopman model is built to characterize the train dynamics. Then, a model predictive controller is designed based on the Koopman model under comfort and actuator constraints. Furthermore, an online update mechanism for the Koopman model is developed to cope with the changing dynamic characteristics of trains, which reduces the accumulation errors and improves control performance. Stability analysis of the closed-loop control system is provided. Comparative simulation results validate the effectiveness of the proposed control approach.

Energy-Efficient Speed Profile Optimization and Sliding Mode Speed Tracking for Metros

Nowadays, most metro vehicles are equipped with an automatic train operation (ATO) system, and the speed control method, combining cruise speed planning and proportional-integral-derivative (PID) control, is widely used. The automation is achieved, and the energy-efficient can be improved. This paper presents an improved artificial bee colony algorithm for speed profile optimization with coast mode and an adaptive terminal sliding mode method for speed tracking. Specifically, a multi-objective optimization model is established, which considers energy consumption, comfortableness, and punctuality. Then, a novel artificial bee colony algorithm named regional reinforcement artificial bee colony (RR-ABC) is designed, to search the optimal speed profile with coast mode, in which some improvements are made to speed up convergence and to avoid local optimal solutions. For speed-tracking control, the adaptive terminal sliding mode controller (ATSMC) is used to improve the speed error, robustness, and energy saving. In addition, a disturbance observer (DOB) is designed to improve the anti-interference ability of the system and further improve the robustness and anti-disturbance, which are also conducive to speed error and energy saving. Finally, the line and train data of the Qingdao Metro Line 6 are used for simulation, which proves the effectiveness of the study. Specific to the energy saving rate, and compared with normal algorithms, RR-ABC with coast mode is approximately 9.55%, and ATSMC+DOB is 7.58%.

Braking process identification of high-speed trains for automatic train stop control

Automatic train stop control (ATSC) is a key function of the automatic train operation (ATO) system. An accurate braking process model can help to improve the control strategy. In this paper, the braking process for stop control of high-speed trains is formulated as a single-point time delay model, based on the principle of practical braking processes. Furthermore, a Picard iteration based identification method is first applied to the time delay system, and a train braking process identification method is proposed. The method is straightforward, and the parameters can be identified based on the principle of ordinary differential equations. The effectiveness of the braking process model and the identification method is illustrated by real-life experimental data.

An Eco-Driving Algorithm for Interoperable Automatic Train Operation

The new Automatic Train Operation (ATO) system over the standard European Rail Traffic Management System (ERTMS) will specify the requirements that an automatic train driving system must fulfil in order to be interoperable. The driving is defined by target times located along the journey that are received from the trackside system. Then, the on-board equipment drives the train with the objective of meeting all of the target times. The use of eco-driving methods to calculate the train driving is necessary, as one of the main goals of modern train driving systems is to increase the energy efficiency. This paper presents a simulation-based optimisation algorithm to solve the eco-driving problem constrained by multiple target times. This problem aims to minimize the energy consumption subject to a commercial running time, as the classical eco-driving problem, and also to meet intermediate target times during the journey between stations to enable automatic traffic regulation, especially at junctions. The algorithm proposed combines a Differential Evolution procedure to generate possible solutions with a detailed train simulation model to evaluate them. The use of this algorithm makes possible to find accurate speed profiles that meet the requirements of multiple time objectives. The proposed Differential Evolution algorithm is capable of finding the feasible speed profile with the minimum energy consumption, obtaining a 7.7% of energy variation in the case of a journey with one intermediate target time and 3.1% in the case of two intermediate targets.

Smart Train Operation Algorithms Based on Expert Knowledge and Reinforcement Learning

During decades, the automatic train operation (ATO) system has been gradually adopted in many subway systems for its low-cost and intelligence. This article proposes two smart train operation (STO) algorithms by integrating the expert knowledge with reinforcement learning algorithms. Compared with previous works, the proposed algorithms can realize the control of continuous action for the subway system and optimize multiple critical objectives without using an offline speed profile. First, through learning historical data of experienced subway drivers, we extract the expert knowledge rules and build inference methods to guarantee the riding comfort, the punctuality, and the safety of the subway system. Then we develop two algorithms for optimizing the energy efficiency of train operation. One is the STO algorithm based on deep deterministic policy gradient named (STOD) and the other is the STO algorithm based on normalized advantage function (STON). Finally, we verify the performance of proposed algorithms via some numerical simulations with the real field data from the Yizhuang Line of the Beijing Subway and illustrate that the developed STO algorithm are better than expert manual driving and existing ATO algorithms in terms of energy efficiency. Moreover, STOD and STON can adapt to different trip times and different resistance conditions.

NSGA-II-Based Parameter Tuning Method and GM(1,1)-Based Development of Fuzzy Immune PID Controller for Automatic Train Operation System

Automatic train operation (ATO) system is one of the important components in advanced train operation control systems. Ideal controllers are expected for the automatic driving function of ATO systems. Aiming at the intelligence requirements of the systems, an NSGA-II-based parameter tuning method for the fuzzy immune PID (FI-PID) controller and a grey model GM(1,1)-based fuzzy grey immune PID (FGI-PID) controller were proposed. Taking a maglev train’s model as the control object and a velocity-time curve as the input, the feasibility of the parameter tuning method for the FI-PID controller and the applicability of the FI-PID controller and the FGI-PID controller for the ATO system were tested. The results showed that the optimized parameters were ideal, the two controllers all showed good performance on the indicators of traceability and comfort level, and the FGI-PID controller performed better than the FI-PID controller. The results exhibited the effectiveness of the proposed methods.