Automatic Train Research Articles

Train-Centric Communication Based Autonomous Train Control System

Driverless train control systems are becoming popular in the world. Notably, this technology has already been applied for several years. However, these kinds of Automatic Train Operation systems are still under the framework that network elements and movement authority are generated from the ground control center. Hence, the aforementioned rail systems are “automated” but not “autonomous,” and the architecture utilization and train tracking interval are limited. With the development of advanced sensors and control algorithms, the train intends to be equipped with a decision-making capability. In order to further improve train movement coordination and efficiency, an Autonomous Train Control System (ATCS) is proposed. This paper provides three main contributions: Firstly, the system structure is described, and the information following in ATCS is discussed with general operation scenarios. Secondly, data isolation is broken under ATCS, and data prediction and edge-based information fusion are applied to process the real dynamic data and estimate the train's speed and position. Thirdly, with the assistant of data fusion, model predictive and data-based control methodologies are discussed.

Residual risk in Automatic Train Protection systems: evaluation and management

The paper on typical rail operation hazards, such as Signal Passed at Danger (SPAD) and Failed Braking Application by Driver (FBAD). The protection against these hazards is normally by Automatic Train Protection (ATP) systems and emergency braking activation, operated by the onboard systems over a certain Minimum Release Speed (MRS). The determination of such speed is a key design parameter related with the achievable performances in terms of safety and capacity. The present research deal with a systematic analysis of the operational conditions potentially affecting the determination of such speed values by modelling the dynamics of the problem taking into account the features of the infrastructures (tracks layout, gradients, etc.), the vehicles (mass, braking performances, etc.) and the geography of signalling and ATP systems themselves. In fact, the position of main signals protecting the dangerous points (switches, level crossings, etc.) emerges as key player, affecting both the design and the operation of the systems.

Deep learning-based bubble detection with automatic training data generation: Application to the PEM water electrolysis

Various fields, such as the paper industry, chemical engineering, and renewable energy, are faced with gas-liquid two-phase flows and are being studied by visualization and observation. Although it is necessary to quantitatively evaluate the characteristics of bubbles, there is a limitation in the amount of labor required for detection and measurement by human observation of images. There are no examples for bubbles in polymer electrolyte membrane water electrolysis (PEMWE), where the bubbles in PEMWE have heterogeneous backgrounds, unpatterned patterns, and unclear bubble contours. Existing methods for detecting these bubbles are not expected to be accurate enough. In this study, a deep learning-based bubble detection method using convolutional neural networks (CNN) was developed for bubbles in PEMWE. Our method has two novel approaches: first, we developed an algorithm that automatically draws a pseudo-bubble image based on an actual bubble image on an arbitrary background as a method for automatically generating training data for the CNNs. Second, we developed a CNN-based bubble detection method that uses the motion of bubbles, specifically, the difference between the bubble image and the image from one frame ago, as input. Finally, we tested the algorithm on bubble images in a PEMWE and achieved an F1 score of 0.83 or better for all current densities of 0.5, 1, 2, and 3 A/cm2, and an F1 score of 0.844 for the entire detection.

Real Time Control System for Metro Railways Using PLC & SCADA

This paper proposes to adopt SCADA and PLC technology for the improvement of the performance of real time signaling & train control systems in metro railways. The main concern of this paper is to minimize the failure in automated metro railways system operator and integrate the information coming from Operational Control Centre (OCC), traction SCADA system, traction power control, and power supply system. This work presents a simulated prototype of an automated metro train system operator that uses PLC and SCADA for the real time monitoring and control of the metro railway systems. Here, SCADA is used for the visualization of an automated process operation and then the whole operation is regulated with the help of PLC. The PLC used in this process is OMRON (NX1P2-9024DT1) and OMRON’s Sysmac studio programming software is used for developing the ladder logic of PLC. The metro railways system has deployed infrastructure based on SCADA from the power supply system, and each station’s traction power control is connected to the OCC remotely which commands all of the stations and has the highest command priority. An alarm is triggered in the event of an emergency or system congestion. This proposed system overcomes the drawbacks of the current centralized automatic train control (CATC) system. This system provides prominent benefits like augmenting services which may enhance a network’s full load capacity and network flexibility, which help in easy modification in the existing program at any time.

Research on ATO Control Method for Urban Rail Based on Deep Reinforcement Learning

Aiming at the problems of punctuality, parking accuracy and energy saving of urban rail train operation, an intelligent control method for automatic train operation (ATO) based on deep Q network (DQN) is proposed. The train dynamics model is established under the condition of satisfying the safety principle and various constraints of automatic driving of urban rail train. Considering the transformation rules and sequences of working conditions between train stations, the agent in the DQN algorithm is used as the train controller to adjust the train automatic driving strategy in real time according to the train operating state and operating environment, and optimizes the generation of the train automatic driving curve. Taking the Beijing Yizhuang Subway line as an example, the simulation test results show that the DQN urban rail train control method reduces energy consumption by 12.32% compared with the traditional train PID control method, and improves the running punctuality and parking accuracy; at the same time, the DQN train automatically driving control method can adjust the train running state in real time and dynamically, and has good adaptability and robustness to the change of train running environment parameters.

AUTOMATIC TRAINING EXPERT SYSTEM

Group training in general secondary education, secondary and higher specialized educational institutions involves the transfer of knowledge and the development of skills, with an orientation to the "average" according to the abilities of the student. At the same time, achieving such a level of training as "skills" is, in principle, impossible. The level of skills requires, as a rule, individual training with the use of various kinds of teaching and training systems together with the method of iterative learning. The coronavirus pandemic (COVID-19) and the transition to distance learning have intensified the development of information and communication technologies (ICT) for teaching without the presence of a teacher. All this determines the relevance of the development of an automatic training system based on an expert system that ensures, in the absence of a teacher, the achievement of the maximum for each student, depending on his intelligence, the required level of training. In this paper, the structure of an automatic learning system is proposed, which ensures the assimilation of educational material to a given level of learning without the participation of a teacher. Using the example of using such a system for training operators of an arc steelmaking furnace, the work of the system elements is shown: the object simulator model, the trainee error analysis blocks, the synthesis of training information, as well as database and knowledge functions. The issues of obtaining andformalizing the knowledge of expert teachers are considered separately. Examples of the formation of knowledge frames for the training of chipboard steelworkers are given. For example, to teach the process of mixing an electric steelmaking process, 81 frames of information are needed, providing educational comments for all kinds of mistakes of the trainees. Conclusions were drawn based on the results of the work.

Multi-objective train speed profile determination for automatic train operation with conscious search: A new optimization algorithm, a comprehensive study

Train speed profile is one of the most fundamental optimization issues in rail transportation. The speed profile has several indices, such as travel time and energy consumption. These indices mostly have opposite natures, which leads to a non-linear, complex multi-objective optimization problem. This paper introduces a new heuristic algorithm called Conscious Search (CS) to determine the optimal speed profile. For this purpose, after modeling the train dynamics, we initially analyze the validated heuristic optimization algorithms, including Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC), Bees Algorithm (BA), and Teaching–learning Based Optimization (TLBO) statistically using sensitivity analysis. Then, using new criteria called the Impact Factor, we show that a trade-off prevails in the global and local search for algorithms. In general, the stronger the algorithm in one domain, the weaker in the other. Subsequently, no algorithm is superior to another. In this regard, CS is proposed as a hybrid algorithm that performs the optimization in two separate global and local phases to address this problem and achieve the ideal optimization. CS steps are designed to dominate other algorithms in both local and global phases. In the evaluation, we obtained the train speed profile using the proposed method and the proven latest versions of the mentioned algorithms. Based on the simulation results, CS outperforms the considered algorithms by a palpable margin, and its solutions dominate others. These results indicate on effectiveness and excellence of the proposed method.

RV_paper

The article presents an overview of the construction of automatic couplers and their digital version of the DAC for freight wagons. In the article, particular attention was paid to the process related to the development and popularization of the design of automatic digital couplers in the European Union countries. Poland, the automatic coupler train was presented in July 2022 and tested under various conditions. This is related to the current trend of improving rail safety and accelerating the procedure of forming trains, and, consequently, sending a freight train on its way. Automatic digital couplers will allow, in addition to mechanical connection, also pneumatic and electrical connection without the participation of an employee for coupling and uncoupling carriages.

Automatic generation of artificial images of leukocytes and leukemic cells using generative adversarial networks (syntheticcellgan)

Background and objectives: Visual analysis of cell morphology has an important role in the diagnosis of hematological diseases. Morphological cell recognition is a challenge that requires experience and in-depth review by clinical pathologists. Within the new trend of introducing computer-aided diagnostic tools in laboratory medicine, models based on deep learning are being developed for the automatic identification of different types of cells in peripheral blood. In general, well-annotated large image sets are needed to train the models to reach a desired classification performance. This is especially relevant when it comes to discerning between cell images in which morphological differences are subtle and when it comes to low prevalent diseases with the consequent difficulty in collecting cell images. The objective of this work is to develop, train and validate SyntheticCellGAN (SCG), a new system for the automatic generation of artificial images of white blood cells, maintaining morphological characteristics very close to real cells found in practice in clinical laboratories. Methods: SCG is designed with two sequential generative adversarial networks. First, a Wasserstein structure is used to transform random noise vectors into low resolution images of basic mononuclear cells. Second, the concept of image-to-image translation is used to build specific models that transform the basic images into high-resolution final images with the realistic morphology of each cell type target: 1) the five groups of normal leukocytes (lymphocytes, monocytes, eosinophils, neutrophils and basophils); 2) atypical promyelocytes and hairy cells, which are two relevant cell types of complex morphology with low abundance in blood smears. Results: The images of the SCG system are evaluated with four experimental tests. In the first test we evaluated the generated images with quantitative metrics for GANs. In the second test, morphological verification of the artificial images is performed by expert clinical pathologists with 100% accuracy. In the third test, two classifiers based on convolutional neural networks (CNN) previously trained with images of real cells are used. Two sets of artificial images of the SCG system are classified with an accuracy of 95.36% and 94%, respectively. In the fourth test, three CNN classifiers are trained with artificial images of the SCG system. Real cells are identified with an accuracy ranging from 87.7% to 100%. Conclusions: The SCG system has proven effective in creating images of all normal leukocytes and two low-prevalence cell classes associated with diseases such as acute promyelocyte leukemia and hairy cell leukemia. Once trained, the system requires low computational cost and can help augment high-quality image datasets to improve automatic recognition model training for clinical laboratory practice.

Automatic model training under restrictive time constraints

We develop a hyperparameter optimisation algorithm, Automated Budget Constrained Training, which balances the quality of a model with the computational cost required to tune it. The relationship between hyperparameters, model quality and computational cost must be learnt and this learning is incorporated directly into the optimisation problem. At each training epoch, the algorithm decides whether to terminate or continue training, and, in the latter case, what values of hyperparameters to use. This decision weighs optimally potential improvements in the quality with the additional training time and the uncertainty about the learnt quantities. The performance of our algorithm is verified on a number of machine learning problems encompassing random forests and neural networks. Our approach is rooted in the theory of Markov decision processes with partial information and we develop a numerical method to compute the value function and an optimal strategy.

Strict rule-based automatic training data extraction using Mobile laser scanning in urban area

To reduce the cost of manually annotating training data for supervised classifiers, we propose an automated approach to extract training data of urban objects in six classes: buildings, fences, man-made poles, vegetation, vehicles, and low objects. In this study, two segmentation algorithms are firstly implemented to generate meaningful objects from the non-ground point cloud. Then, we generated valid strict rules to label partial RANSAC (Random Sample Consensus) planes and meaningful objects as training data. The strict rules are built upon the semantic knowledge formed by the features of geometric, eigenvalue, RANSAC plane, multidimensional slice, and relative location. The accuracy of strict rule-based (SRB) training data is higher than 98.5 % for buildings, man-made poles, vegetation, and vehicles. The accuracy of low objects and fences reaches 97.10 % and 94.99 %, respectively. Finally, we compared the performance of the KPConv and PointNET++ networks trained by SRB and manually labeled training data to evaluate the effectiveness of our training data. The KPConv overall accuracy using manually labeled and (SRB) training data are 91.5 % and 86.8 % in the Paris dataset, 95.6 % and 92.0 % in the Freiburg dataset, respectively. The experiments demonstrate that automatically labeled training data can achieve similar accuracy compared to manual labels when coupled with two deep learning networks. Therefore, SRB training data extraction can effectively deal with the problem of training data scarcity and provide significant advancements in urban point cloud classification, where manual labeling of training data remains a crucial challenge.

МЕТОД ОПТИМИЗАЦИИ КЛАССИФИКАТОРА ЖЕСТОВ

The work is devoted to the study of the possibility of optimizing the process of synthesis ofgesture classifiers by selecting the most significant channels of electromyographic (EMG) activityof the muscles of the forearm. The first part of the study is devoted to the development and analysisof the performance of gesture classifiers with a different number of EMG channels, ranked bysignificance based on the Pearson criterion. The solution of the problem of classification of gesturesby EMG signals was first implemented on the basis of ensembles of decision trees trained bythe gradient boosting method. For this, software was developed that allows automatic synthesisand training of gesture classifiers. Next, a series of studies was carried out to find the optimalnumber of EMG channels based on three criteria: the classifier learning rate, the performance ofthe trained model, and the area under the ROC AUC error curve. To do this, a cycle of trainingand testing of the classifier was carried out for data sets recorded at different positions of the electrodeson the forearm. Then, range diagrams of the studied criteria were constructed for variousnumbers of EMG channels involved in the work from 1 to 8, ranked by significance in each of thesamples. It was found that the optimal number of EMG channels involved under the experimentalconditions was 3-6, since a further increase did not lead to a decrease in the classification error,while significantly degrading the performance. The proposed method allows you to automaticallyselect the channels, the electrodes of which are located above the most informative areas of theforearm in case of an accidental change in the position of the sensors. The second part of the workcontains the results of a full-scale experiment to demonstrate the possibility of controlling awheeled robot through EMG analysis.

Design of Redundant Manipulator System for Water Injection of EMU Train

Robots are known as "the Pearl at the top of the crown of manufacturing." As the most common kind of robot, the multi-degree-of-freedom manipulator has its reliability, lightweight, integration, and modularization, which are the focus of researchers. Based on the background of injecting water into the EMU train, this paper uses the water feeding manipulator (WFM) to completely replace the manual operation and researches the key technology of the redundant manipulator system. At the component level, a new type of planetary thread roller bearing (PTRB) is used as the main bearing component of the manipulators’ joint. For joint, two integrated joint modules (IJM) are developed based on PTRB to be an option for IJM with a high power-weight ratio and high reliability. Finally, for the system, according to the fault-tolerant performance requirements of the task point, the redundant manipulator system architecture is proposed, the mechanical body is set as the redundant structure, the driver and controller are dynamically reconstructed, and the control algorithm and fault diagnosis are fault-tolerant. The failure rate is 3.257×10-8/h, laying the foundation for the realization of a fully automatic and intelligent train water supply.

Cuckoo search approach for automatic train regulation under capacity limitation

Cuckoo search approach for automatic train regulation under capacity limitation

Accurate Parking Control for Urban Rail Trains via Robust Adaptive Backstepping Approach

Nowadays, precise train parking has become a key technology of automatic train operation. Due to the high nonlinearities and various uncertainties existing in the dynamics of urban rail trains (URTs), it is challenging to develop an effective parking controller to achieve high-precision parking. To address this issue, a novel robust adaptive backstepping controller is proposed in the paper, where a robust term is employed to compensate for the system uncertainties caused by the brake shoe, and meanwhile several parametric adaption laws are equipped to estimate the uncertain system parameters as well as the external disturbances. The closed-loop stability of the controlled system is analyzed rigorously by virtue of the Lyapunov theory, and the effectiveness of the proposed control scheme in precise parking of URTs is demonstrated through numerical simulations.

A Comprehensive Resilient Control Strategy for CBTC Systems Through Train-to-Train Communications Under Malicious Attacks

In recent years, intelligent transportation systems and automatic train control technologies have been developed rapidly. The communication-based train control (CBTC) system through train-to-train communications is very important in the intelligent transportation system and it uses wireless communication links to transmit the train control commands. However, the CBTC system is sensitive to information security attacks. Therefore, a comprehensive resilient control strategy is proposed for the CBTC system under Denial of Service (DoS) attacks. As a cyber physical system, the CBTC system includes two layers. In the communication layer, a strongly robust train-to-train communication topology is proposed which adopts a bidirectional communication strategy and a three times communication protocol to achieve the goal of resisting multi-channel DoS attacks. In the physical layer, an observer-based distributed intrusion detection method is firstly proposed to detect the attack. Further, a resilient control method based on a distributed state observer of the leader train is proposed to avoid triggering the train safety protection mechanism frequently. Finally, simulation results prove that the proposed comprehensive resilient control algorithm is effective in the CBTC system.

OBJECT CLASSIFICATION BASED ON THERMAL IMAGES USING DEEP LEARNING (PRE-TRAINED RESNET 50)

The development of technology in the field of transportation is currently something that is very enthusiastically welcomed by the Indonesian people in general. But along with the development of transportation technology that exists today is much different from the past, with a lot of sophistication and improving quality and safety that is more innovative. An Autonomous Car was created that can make it easier for drivers and maintain safety while driving. This system was built using the Neural Network control method, as well as Image Processing as the input signal in the form of images, and with the Flir camera as the vehicle data input. This of course has a very positive impact on human life today, of course humans will be more efficient in time, maintain safety on the trip, and can be more productive when driving. The method that is currently developing rapidly is automatic extraction using deep learning technology. The method used is Fully Convolutional Network (FCN) with Residual Neural Network (ResNet) architecture.The method currently used in the research is automatic extraction using deep learning technology to detect objects in the classification that has been made, with Residual Neural Network 50 (ResNet) architecture. In this study, the extraction accuracy for automatic vehicle function training reached 97.1% for ResNet 50 and 96.7% for ResNet 101 with a resolution of 640x512 pixels.

Peningkatan Kualitas Kemasan Produk menggunakan Mesin Pres Segel Plastik Otomatis di Komunitas UMKM Shingkara Kecamatan Singosari Kabupaten Malang

Micro, Small, and Medium Enterprises (UMKM) are one of the driving forces of the economy in a local area as well as a booster for the community's welfare. One of the obstacles often faced by the Shingkara UMKM community in Singosari District, Malang Regency is the lack of quality product packaging that still uses manual tools and is not optimal in creating marketing content. This community service aims to provide automatic plastic seal press machines and training on marketing content creation for micro-business actors in the Shingkara UMKM community so that it is expected to increase partners' competitiveness. The stages of community service were observation and coordination, providing machines, handing over machines, holding training, and evaluation. The results of the community service activities showed that the training participants were enthusiastic and actively participated during the training. The quality of product packaging after using a plastic seal press machine has also improved, evidenced by a tighter and well-protected packaging seal. The copywriting training has also broadened the participants' knowledge about strategies for creating product marketing content by utilizing applications available on Android.

Energy-efficient automatic train operation for high-speed railways: Considering discrete notches and neutral sections

The energy-efficient driving of automatic train operation (ATO) system in high-speed railway systems has attracted more and more attention. Different from the energy-efficient automatic train control problem for metro trains, the traction/braking forces given by high-speed ATO systems are currently not continuous but discrete values corresponding to traction/braking notches due to the limited communication interface between ATO systems and rolling stocks. In addition, since high-speed trains are powered by single-phase alternating current, there is no power supply for train traction in neutral sections, which is introduced to separate two catenaries powered by currents with different phases. In this paper, we propose a discrete-space based mathematical formulation for the energy-efficient automatic train control problem for high-speed trains, where binary variables are introduced to indicate the traction/braking notch selections and to indicate whether a discrete space interval is a neutral section or not. The mathematical formulation is then transformed into a mixed integer linear programming (MILP) model, which can be solved by existing solvers. However, due to the discretization of the space the discrete traction/braking notches, the resulting MILP problem could be infeasible and the optimal driving strategy cannot be obtained. Hence, we present a novel relaxation method, i.e., the final braking notch relaxation method, to relax a few constraints in the final braking phase and to generate a feasible speed profile. Comprehensive experiments are investigated based on Beijing–Harbin high-speed railway line, where the energy consumption of the speed profiles with discrete (i.e., notch-based) traction/braking forces is slightly higher than that with continuous traction/braking forces. Moreover, the impacts of the neutral section locations and the effects of different numbers of traction/braking notches are deeply analyzed.

Fuzzy Adaptive Protective Control for High-Speed Trains: An Outstretched Error Feedback Approach

This paper presents a fuzzy adaptive protective control method for autonomous high-speed trains (HSTs) automatic operation using a new outstretched error feedback design approach. In order to stabilizing the error dynamics with respect to target position and speed profiles of controlled HSTs, nonlinear transformation in prescribed performance control methodology is used to convert running states subject to protective (constrained) information, imposed by automatic train protection (ATP) subsystem, to new coordinates in unconstrained form. By blending a new outstretched error feedback and fuzzy approximation, it is guaranteed above-mentioned errors are kept within regions characterized by error boundary (or prescribed performance) functions and control parameters simultaneously, which can be adjusted to arbitrarily small even without error decreasing boundary functions. Fuzzy approximation is used in compensating unknown running resistances. It is rigorously proved that the resulting closed-loop system is stable in sense of Lyapunov stability in the presence of unknown resistance, containing basis and aerodynamic resistances with uncertain parameters and piecewise continuously slope resistance over varying gradient profile. The proposed control is demonstrated to be effective by comparative simulations of train G1 running on Beijing-Shanghai railway line.