• Journal of the Korean Society for Railway
• /
• v.11 no.2
• /
• pp.145-150
• /
• 2008

The objective of this research is to design the roller rig for the maintenances of high-speed train bogies operated on the tracks. Roller rigs have been studied and researched to develop the faster, safer and more efficient railway system. It is to reduce the time of testing vehicles and to make as wide a range of tests available as possible. Therefore, it is very important issue to check and evaluate the dynamic responses of high-speed train after several years of operation. This paper presents a study on the conceptual design and dynamic model to develop the roller rig for the routine maintenances of high-speed trains bogies with maximum speed of 350km/h. ANSYS was used to analyze the wheel/roller's contact behavior of driving axle and ADAMS was used to verify and analyze the dynamic behaviors of roller rig.

• The Journal of the Korea Contents Association
• /
• v.17 no.1
• /
• pp.50-57
• /
• 2017

This study investigated the causes and problems of train operation impediments through the statistics analysis of 8 years'internal data of Daejeon Metropolitan Express Transit. By evaluating the risks regarding the system, equipment, and parts of high risk group, this study measured the Risk Index Severity, and applied the $5{\times}5$ Risk Assessment Matrix which is a method of risk management to calculate the scale of risk to analyze the safety level and allowance range. As a result, the car sector, the most serious risk, followed by machinery and equipment sector showed that the inherent risk. In particular, the door broken and the door rail signaling and control devices due to defects of the vehicle is high, but also the severity, and frequency are showing very frequent additional potential accidents. PSD also had defects in the machinery sector appeared to be the most dangerous of the PSD poor safety gates, it was found that the glass also involve the risk of mishandling and breakage of the PSD. This study intended to contribute to the transportation benefits through the safety and stable operation of Metropolitan Express Transit.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.12 no.10
• /
• pp.758-765
• /
• 2002

KITECH and ODS performed a study of internal and external noise prediction of the Korean high speed prototype test train(HSR 350X). The object of this study was 3 kinds of cars, trailer car(TT2), motorized car(TMI ) and power car(TPI) and the predicted noise was for the two different driving speeds in free field and tunnel conditions. Data of carbody design and noise sources were delivered from manufactures. Some of noise sources which were not available in the project team, were chosen by experiences of ODS. Internal noise level of each car was predicted for two cases i.e, at 300 km/h and 350 km/h. In addition sound transmission path and dominant noise sources were also investigated for each section of the car, which is circular shell typed part of whole carbody. In case of TT2, the dominating sound transmission path is the (floor in terms of structure-borne noise and air-borne noise. The main noise sources are structure-borne noise from the yaw-damper and air-borne noise from the wheel/rail contact, whereas the dominating sound transmission path of TMI are floor and sidewall below the window in terms of structure-borne noise. The main noise sources of TMI are structure-borne noise from motor/gear unit and the yaw-damper in the free field, and air-borne noise from the wheel/rail contact and structure-borne noise from motor/gear unit in the tunnel. Through the external noise prediction for the KHST test train formation, the noise form the wheel/rail contact is estimated as one of the major sources. In addition, the noise specification of sub-component was proposed for managing each sub-surpplier to reach the KHST noise requirement. The specification provide the sound power of machinery part and transmission loss of component of carbody structure. The predicted noise level in each case exceeded the required limit. Through this study, the noise characteristics of the test train were investigated by simulation, and then the actual test will be performed in near future. Both measured and calculated data will be compared and further work for noise reduction will be continued.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.1
• /
• pp.19-25
• /
• 2011

A vehicle fuel economy is very important issue in view of fuel cost and environmental regulation. The technology development for the fuel economy improvement improved the engine, power train and many components of vehicle. So, the fuel economy is much improved, but up to now the measurement of it tests the given mode(LA-4, FTP-75, etc) within computer simulation program and engine dynamo. In this paper, to deduct the method of its improvement of real road, the test vehicle ran 213Km Youngdong real highway using 3 different algorithms in computer simulation. For this, I extracted the distance and altitude data from received GPS data and calculated the grade angle, road load and accomplished the velocity profiles according to algorithms in all 213Km distance. The vehicle runs in computer with AVL Cruise simulation program using velocity profile. I calculate the fuel economy using AVL Cruise simulation result and propose the Eco-driving method of them.

• Journal of Broadcast Engineering
• /
• v.28 no.3
• /
• pp.302-313
• /
• 2023

This paper aims to develop a framework that can fully automate the quality management of training data used in large-scale Artificial Intelligence (AI) models built by the Ministry of Science and ICT (MSIT) in the 'AI Hub Data Dam' project, which has invested more than 1 trillion won since 2017. Autonomous driving technology using AI has achieved excellent performance through many studies, but it requires a large amount of high-quality data to train the model. Moreover, it is still difficult for humans to directly inspect the processed data and prove it is valid, and a model trained with erroneous data can cause fatal problems in real life. This paper presents a dataset reconstruction framework that removes abnormal data from the constructed dataset and introduces strategies to improve the performance of AI models by reconstructing them into a reliable dataset to increase the efficiency of model training. The framework's validity was verified through an experiment on the autonomous driving dataset published through the AI Hub of the National Information Society Agency (NIA). As a result, it was confirmed that it could be rebuilt as a reliable dataset from which abnormal data has been removed.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.12
• /
• pp.3330-3344
• /
• 2023

This paper proposes an antenna performance prediction model in the autonomous driving radar manufacturing process. Our research work is based upon a challenge dataset, Driving Radar Manufacturing Process Dataset, and a typical AutoML machine learning workflow engine, Pycaret open-source Python library. Note that the dataset contains the total 70 data-items, out of which 54 used as input features and 16 used as output features, and the dataset is properly built into resolving the multi-output regression problem. During the data regression analysis and preprocessing phase, we identified several input features having similar correlations and so detached some of those input features, which may become a serious cause of the multicollinearity problem that affect the overall model performance. In the training phase, we train each of output-feature regression models by using the AutoML approach. Next, we selected the top 5 models showing the higher performances in the AutoML result reports and applied the ensemble method so as for the selected models' performances to be improved. In performing the experimental performance evaluation of the regression prediction model, we particularly used two metrics, MAE and RMSE, and the results of which were 0.6928 and 1.2065, respectively. Additionally, we carried out a series of experiments to verify the proposed model's performance by comparing with other existing models' performances. In conclusion, we enhance accuracy for safer autonomous vehicles, reduces manufacturing costs through AutoML-Pycaret and machine learning ensembled model, and prevents the production of faulty radar systems, conserving resources. Ultimately, the proposed model holds significant promise not only for antenna performance but also for improving manufacturing quality and advancing radar systems in autonomous vehicles.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.16 no.4
• /
• pp.153-163
• /
• 2017

This research is based on the previous research that personally preferred safe distance, rotating angle and speed are differentiated. Thus, we use machine learning model for recognizing maneuvering modes trained per personal or per similar driving pattern groups, and we evaluate automatic driving according to maneuvering modes. By utilizing driving knowledge, we subdivided 8 kinds of longitudinal modes and 4 kinds of lateral modes, and by combining the longitudinal and lateral modes, we build 21 kinds of maneuvering modes. we train the labeled data set per time stamp through RNN, LSTM and Bi-LSTM models by the trips of drivers, which are supervised deep learning models, and evaluate the maneuvering modes of automatic driving for the test data set. The evaluation dataset is aggregated of living trips of 3,000 populations by VTTI in USA for 3 years and we use 1500 trips of 22 people and training, validation and test dataset ratio is 80%, 10% and 10%, respectively. For recognizing longitudinal 8 kinds of maneuvering modes, RNN achieves better accuracy compared to LSTM, Bi-LSTM. However, Bi-LSTM improves the accuracy in recognizing 21 kinds of longitudinal and lateral maneuvering modes in comparison with RNN and LSTM as 1.54% and 0.47%, respectively.

• Proceedings of the KSR Conference
• /
• 2011.05a
• /
• pp.918-923
• /
• 2011

On AC railway systems, the neutral sections are installed in front of substations and sectioning posts in order to avoid crash between power that have differing phases. In case railway vehicles pass through these neutral sections, it is necessary for them to switch to coasting driving by notch-off. This may reduce speed of the vehicles, resulting lowered train operation efficiency. The usage of automatic power switching systems makes it possible to pass neutral sections at notch-on, enhancing operation efficiency so that it is appropriate for high-speed railway applications. This paper introduces a simulator that assesses efficiency of automatic power switching systems in neutral sections. The is composed of a power supply system, electric railway vehicles, thyristor switches, and traction motors.

• Proceedings of the KSR Conference
• /
• 2011.10a
• /
• pp.2866-2869
• /
• 2011

Concrete track to increase R&D compared to the existing gravel track 3dB(A) over the growing problem of noise has been raised. Accordingly, the noise reduction solutions for reducing the vibration of the rail that you want to reduce the noise of the concept is to develop the rail web-damper. For this purpose, first, that occurs while driving the train to simulate the vibration modes of rail vibration part of the main draw for this part of the effective vibration reduction to be made, a review of various shapes to try.

• Proceedings of the KSR Conference
• /
• 2011.10a
• /
• pp.1888-1893
• /
• 2011

Considering the dynamic performance and stability of railroad vehicles has begun to grab the attention because of developing the high speed train recently. A development based on an analysis of dynamics and verification has to be required to study the stability of vehicle performance. Several ways of analysis were using the look-up table to apply the wheel-rail contact characteristics quickly, whereas there is a constraint of the wheelset lateral displacement. In this study, an development of searching the wheel-rail contact position has been provided. The 2-point contact between wheel and rail during the driving condition can be calculated by numerical analysis. Moreover, a reliability is verified by comparing the result with a commercial program.