• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.6
• /
• pp.1337-1345
• /
• 2009

In the railway, the delay of a train may affect the schedule of other trains. Hence, the timetable management activity involving overtaking and transferring becomes an important issue. The timetable rescheduling is the conflict resolution of the disrupted schedule and a matter of deciding the sequence of moves among trains and the place for meeting. In Korea, there are few cases where operations research models were applied to railway timetable rescheduling problem in consideration of the transfer between Korea Train eXpress (KTX) trains and conventional trains. Hence, we present a mathematical approach that can minimize the total delay of the whole trains secondly. We applied the model to the exemplary section of a double-line track, the Gyung-Boo Line, and it is confirmed that the mathematical model could effectively address the transfer service as well as the complicated railway conflicts of Gyung-Boo Line.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.10a
• /
• pp.211-217
• /
• 1998

Many practical control problems for the complex, uncertain or large-scale plants, need to simultaneously achieve a number of objectives, which may conflict or compete with each other. If the conventional optimization methods are applied to solve these control problems, the solution process may be time-consuming and the resulting solution would ofter lose its original meaning of optimality. Nevertheless, the human operators usually performs satisfactory results based on their qualitative and heuristic knowledge. In this paper, we investigate the control strategies of the human operators, and propose a fuzzy model-based multi-objective satisfactory controller. We also apply it to the automatic train operation(ATO) system for the magnetically levitated vehicles(MAGLEV). One of the human operator's strategies is to predict the control result in order to find the meaningful solution. In this paper, Takagi-Sugeno fuzzy model is used to simulated the prediction procedure. Another str tegy is to evaluate the multiple objectives with respect to their own standards. To realize this strategy, we propose the concept of a satisfactory solution and a satisfactory control scheme. The MAGLEV train is a typical example of the uncertain, complex and large-scale plants. Moreover, the ATO system has to satisfy multiple objectives, such as seed pattern tracking, stop gap accuracy, safety and riding comfort. In this paper, the speed pattern tracking controller and the automatic stop controller of the ATO system is designed based on the proposed control scheme. The effectiveness of the ATO system based on the proposed scheme is shown by the experiments with a rotary test bed and a real MAGLEV train.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.33 no.2
• /
• pp.121-128
• /
• 2020

The purpose of this study is to investigate the distribution patterns of displacement and acceleration fields in a nonlinear soil ground based on the interaction of high-speed train, wheel, rail, and ground. For this purpose, a high-speed train in motion was modeled as the actual wheel, and the vertical contact of wheel and rail and the lateral contact, caused by meandering motion, were simulated; this simulation was based on the moving mass analysis. The soil ground part was given the nonlinear behavior of the upper ground part by using the modified the Drucker-Prager model, and the changes in displacement and acceleration were compared with the behavior of the elastic and inelastic grounds. Using this analysis, the displacement and acceleration ranges close to the actual ground behavior were addressed. Additionally, the von-Mises stress and equivalent plastic strain at the ground were examined. Further, the equivalent plastic and total volumetric strains at each failure surface were examined. The variation in stresses, such as vertical stress, transverse pressure, and longitudinal restraint pressure of wheel-rail contact, with the time history was investigated using moving mass. In the case of nonlinear ground model, the displacement difference obtained based on the train travel is not large when compared to that of the elastic ground model, while the acceleration is caused to generate a large decrease.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.36 no.11
• /
• pp.1447-1453
• /
• 2012

For an electromagnetic suspension (EMS)-type urban Maglev train using U-shaped electromagnets, both the vertical and the lateral air gaps for levitation are maintained only by the electromagnet. The train can run over curved rails without active lateral air gap control because the U-shaped electromagnet simultaneously produces both a levitation force and a guidance force, which is dependent on the levitation force. Owing to the passive control of the lateral air gap, the lateral vibration could exceed the limits of the lateral air gap and acceleration. In this study, dynamic analysis of a Maglev train is carried out, and the effectiveness of a lateral damper for vibration reduction is investigated. To more accurately predict the lateral vibration, a Maglev vehicle multibody model including air-sparing, guideway irregularities, electromagnets, and their controls is developed.

• Journal of the Korean Society for Railway
• /
• v.15 no.5
• /
• pp.467-475
• /
• 2012

Static and dynamic train load tests were conducted to evaluate the train load transfer mechanism in the roadbed which was retained by two types (fully and partially) of segmental retaining walls reinforced by geogrid. The test roadbed was 2.6m high, 5m wide, and 6m long. A combination of earth pressure gages, displacement transducers, and strain gages were placed in specific locations to measure the responses. Test results showed that the wall displacement pattern as well as the earth pressure for the fully reinforced retaining wall was different from those for the partially reinforced retaining wall. In the dynamic train load test, the strain in the upper part of the wall tended to decrease, and both the residual deformation and the rate of the deformation were significantly lower than those in the current design standard.

• Proceedings of the KSR Conference
• /
• 2008.11b
• /
• pp.1169-1177
• /
• 2008

The system of the railway signaling using the track transmits the approved speed to the location of a train and it. Referring to the way of transmitting train control information, there are the one transmitting it to the on-board system of a train using the direct track, the another transmitting it establishing an instrument, and the other transmitting an instrument by a railway track. The one is the method using the direct track as a conductor for composing the part of the track and attaining the information controlling a train by transmitting a signal to the track. It is used for the high-speed railway and the subway. The method using the track attains information by transmitting it to returned information, and the on-board system of a train attains it by magnetic coupling. Because many reinforcing bars on the concrete slab track are used, interaction between a rail and a reinforcing bar that is not produced on ballast track is made. Due to the interaction, the electric characteristic of rail is changed. In the current paper, we numerically computed the coupling coefficient between the rail and the reinforcing bar based on the concrete slab track throughout the model related to the rail and the reinforcing bar using the concrete slab track that is used in the second interval of the Gyeongbu high-speed railway, and we defined the coupling coefficient not changed in the electric characteristic of rail in the condition that there is no interaction between the rail and the reinforcing bar.

• Wind and Structures
• /
• v.35 no.2
• /
• pp.83-92
• /
• 2022

When the rolling stocks run on the curve, the external rail has to be lifted to a certain level to balance the centrifugal force acting on the train body. Under such a situation, passengers may feel uncomfortable, and the slanted vehicle has the potential overturning risks at high speed. This paper conducted a wind tunnel test in an annular wind tunnel with φ=3.2 m based on a 1/20^th scaled high-speed train (HST) model. The sensitivity of Reynolds effects ranging from Re = 0.37×10⁶ to Re = 1.45×106 was tested based on the incoming wind from U=30 m/s to U=113 m/s. The wind speed covers the range from incompressible to compressible. The impact of roll angle ranging from γ=0° to γ=4° on train aerodynamics was tested. In addition, the boundary layer development was also analyzed under different wind speeds. The results indicate that drag and lift aerodynamic coefficients gradually stabilized and converged over U=70 m/s, which could be regeared as the self-similarity region. Similarly, the thickness of the boundary layer on the floor gradually decreased with the wind speed increase, and little changed over U=80 m/s. The rolling moment of the head and tail cars increased with the roll angle from γ=0° to γ=4°. However, the potential overturning risks of the head car are higher than the tail car with the increase of the roll angle. This study is significant in providing a reference for the overturning assessment of HST.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.2
• /
• pp.742-756
• /
• 2022

A flood of information has occurred with the rise of the internet and digital devices in the fourth industrial revolution era. Every millisecond, massive amounts of structured and unstructured data are generated; smartphones, wearable devices, sensors, and self-driving cars are just a few examples of devices that currently generate massive amounts of data in our daily. Machine learning has been considered an approach to support and recognize patterns in data in many areas to provide a convenient way to other sectors, including the healthcare sector, government sector, banks, military sector, and more. However, the conventional machine learning model requires the data owner to upload their information to train the model in one central location to perform the model training. This classical model has caused data owners to worry about the risks of transferring private information because traditional machine learning is required to push their data to the cloud to process the model training. Furthermore, the training of machine learning and deep learning models requires massive computing resources. Thus, many researchers have jumped to a new model known as "Federated Learning". Federated learning is emerging to train Artificial Intelligence models over distributed clients, and it provides secure privacy information to the data owner. Hence, this paper implements Federated Averaging with a Deep Neural Network to classify the handwriting image and protect the sensitive data. Moreover, we compare the centralized machine learning model with federated averaging. The result shows the centralized machine learning model outperforms federated learning in terms of accuracy, but this classical model produces another risk, like privacy concern, due to the data being stored in the data center. The MNIST dataset was used in this experiment.

• Journal of computational fluids engineering
• /
• v.16 no.2
• /
• pp.17-23
• /
• 2011

The turbulent flow and vortex shedding phenomena around pantograph panhead of high speed train were investigated and compared with available experimental data and other simulations. The pantograph head was simplified to be a square-cross-section pillar and assumed to be no interference with other bodies. The Reynolds number (Re) was 22,000. The LES(large eddy simulation) of FDS code was applied to solve the momentum equations and the Wener-Wengle wall model was employed to solve the near wall turbulent flow. Smagorinsky model($C_s$=0.2) was used as SGS(subgrid scale) model. The total grid numbers were about 9 millions and the analyzed domain was divided into 12 multi blocks which were communicated with each other by MPI. The time-averaged mainstream flows were calculated and well compared with experimental data. The phased-averaged quantities had also a good agreement with experimental data. The near-wall turbulence should be carefully treated by wall function or direct resolution to get successful application of LES methods.

• The Journal of the Acoustical Society of Korea
• /
• v.10 no.2
• /
• pp.29-35
• /
• 1991

In this paper, we discuss the method of recognizing allophone unit isolated words using hidden Markov model(HMM). Frist we constructed allophone lexicon by extracting allophones from training data and by training allophone HMMs. And then to recognize isolated words using allophone HMMs, it is necessary to construct word dictionary which contains information of allophone sequence and inter-allophone transition probability. Allophone sequences are represented by allophone HMMs. To see the effects of inter-allophone transition probability and to determine optimal probabilities, we performend some experiments. And we showed that small number of traing data and simple train procedure is needed to train word HMMs of allophone sequences and that not less performance than word unit HMM is obtained.