• Journal of the Ergonomics Society of Korea
• /
• v.22 no.1
• /
• pp.81-94
• /
• 2003

The British railway safety research group has developed a risk assessment model for the railway infrastructure and major railway accidents. The major hazardous factors of the railway infrastructure were identified and classified in the model. The frequency rates of critical top events were predicted by the fault tree analysis method using failure data of the railway system components and ratings of railway maintenance experts, The consequences of critical top events were predicted by the event tree analysis method. They classified the Joss of accident due to railway system into personal. commercial and environmental damages. They also classified 110 hazardous event due to railway system into three categories. train accident. movement accident and non-movement accident. The risk assessment model of the British railway system has been designed to take full account of both the high frequency low consequence type events (events occurring routinely for which there is significant quantity of recorded data) and the low frequency high consequence events (events occurring rarely for which there is little recorded data). The results for each hazardous event were presented in terms of the frequency of occurrence (number of events/year) and the risk (number of equivalent fatalities per year).

• Journal of the Korean Society of Safety
• /
• v.31 no.3
• /
• pp.28-33
• /
• 2016

Since defects in the wheels of railway vehicles, which occur due to wears with the rail, cause serious damage to the running device, the diagnostic monitoring system for condition-based maintenance is required to secure the driving safety. In this paper, we studied to apply a useful Cepstrum analysis to detect periodic structure in spectrum among the vibration signal processing techniques for the fault diagnosis of a rotating body such as wheel. In order to analyze in variations of train velocity, the Cepstrum analysis was performed after a domain change of the vibration signal from time domain to rotation angle domain. When domains change, it is important to use a interpolation for a uniform interval of the rotation angle. Finally, the Cepstrum analysis for wheel flat detection was verified by using the vibration signal including the disturbance resulting from the rail irregularities and the vibration of bogie components.

• Journal of the Korean Society for Railway
• /
• v.18 no.5
• /
• pp.439-446
• /
• 2015

The turnout system, which determines the direction of the train, is not only a key system but also a vulnerable system. Failure of this system may lead to a delay of the train or even casualties. In this light, it is necessary to precisely the conditions of the turnout system. Currently, ROADMASTER of Germany is used as a diagnostic system in Korea. However, a new diagnostic system should be developed for optimized operation of the turnout system with maintenance that is suitable for the Korean railway environment. In this paper, a Fault Tree Analysis for the representative faults of the turnout system is conducted and physical quantities, which can be the cause of the fault, are classified according to the component and function. Also, the measuring factors for the monitoring are derived and a decision making theory is suggested. On the basis of the results, we propose a new turnout diagnostic system that can provide more driverse and precise information than the conventional system.

• Geophysics and Geophysical Exploration
• /
• v.25 no.2
• /
• pp.59-70
• /
• 2022

Recent studies demonstrate that machine learning has expanded in the field of seismic interpretation. Many convolutional neural networks have been developed for seismic sequence identification, which is important for seismic interpretation. However, expense and time limitations indicate that there is insufficient data available to provide a sufficient dataset to train supervised machine learning programs to identify seismic sequences. In this study, patch division and data augmentation are applied to mitigate this lack of data. Furthermore, to obtain spatial information that could be lost during patch division, an artificial channel is added to the original data to indicate depth. Seismic sequence identification is performed using a U-Net network and the Netherlands F3 block dataset from the dGB Open Seismic Repository, which offers datasets for machine learning, and the predicted results are evaluated. The results show that patch-based U-Net seismic sequence identification is improved by data augmentation and the addition of an artificial channel.

• Smart Structures and Systems
• /
• v.29 no.6
• /
• pp.757-766
• /
• 2022

To train deep learning algorithms, a sufficient number of data are required. However, in most engineering systems, the acquisition of fault data is difficult or sometimes not feasible, while normal data are secured. The dearth of data is one of the major challenges to developing deep learning models, and fault diagnosis in particular cannot be made in the absence of fault data. With this context, this paper proposes an anomaly detection methodology for rotating machines using only normal data with self-labeling. Since only normal data are used for anomaly detection, a self-labeling method is used to generate a new labeled dataset. The overall procedure includes the following three steps: (1) transformation of normal data to self-labeled data based on a pretext task, (2) training the convolutional neural networks (CNN), and (3) anomaly detection using defined anomaly score based on the softmax output of the trained CNN. The softmax value of the abnormal sample shows different behavior from the normal softmax values. To verify the proposed method, four case studies were conducted, on the Case Western Reserve University (CWRU) bearing dataset, IEEE PHM 2012 data challenge dataset, PHMAP 2021 data challenge dataset, and laboratory bearing testbed; and the results were compared to those of existing machine learning and deep learning methods. The results showed that the proposed algorithm could detect faults in the bearing testbed and compressor with over 99.7% accuracy. In particular, it was possible to detect not only bearing faults but also structural faults such as unbalance and belt looseness with very high accuracy. Compared with the existing GAN, the autoencoder-based anomaly detection algorithm, the proposed method showed high anomaly detection performance.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.251-266
• /
• 2022

Structural Health Monitoring (SHM) of critical infrastructure comprises a major pillar of maintenance management, shielding public safety and economic sustainability. Although SHM is usually associated with data-driven metrics and thresholds, expert judgement is essential, especially in cases where erroneous predictions can bear casualties or substantial economic loss. Considering that visual inspections are time consuming and potentially subjective, artificial-intelligence tools may be leveraged in order to minimize the inspection effort and provide objective outcomes. In this context, timely detection of sensor malfunctioning is crucial in preventing inaccurate assessment and false alarms. The present work introduces a sensor-fault detection and interpretation framework, based on the well-established support-vector machine scheme for anomaly detection, combined with a coalitional game-theory approach. The proposed framework is implemented in two datasets, provided along the 1st International Project Competition for Structural Health Monitoring (IPC-SHM 2020), comprising acceleration and cable-load measurements from two real cable-stayed bridges. The results demonstrate good predictive performance and highlight the potential for seamless adaption of the algorithm to intrinsically different data domains. For the first time, the term "decision trajectories", originating from the field of cognitive sciences, is introduced and applied in the context of SHM. This provides an intuitive and comprehensive illustration of the impact of individual features, along with an elaboration on feature dependencies that drive individual model predictions. Overall, the proposed framework provides an easy-to-train, application-agnostic and interpretable anomaly detector, which can be integrated into the preprocessing part of various SHM and condition-monitoring applications, offering a first screening of the sensor health prior to further analysis.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.7
• /
• pp.394-399
• /
• 2017

In Korean AC power railway substations, SCOTT winding transformers are under operation to have a single phase power supply together with a phase angle of $90^{\circ}$ on the secondary side of the main transformer. In the case of an internal fault of the transformer, the transformer protection relay should be cut off on the primary side, the transformer should be inoperative to the external fault of the transformer or to the normal train operation. Reducing the malfunction of the relay through an exact fault determination is very important for securing a stable power system and improving its reliability. The main transformers are protected using Buchholtz's relay and a differential relay as the internal fault detection devices, but there are some cases of the main transformer operation under the deactivation of this protection function due to a malfunction of the differential relay. In this paper, the characteristics of the SCOTT transformer and differential relay as well as the malfunctioning of the protection relays are presented. The modeling of the SCOTT transformer protection relay was accomplished by the power system analysis program and the Comtrade file from 'A substation', which was used as the input data for the fault wave, and the harmonics were analyzed to determine if the relay operates or not. In addition, an improvement plan for malfunctioning cases through wave form analysis is suggested.

• Proceedings of the KIPE Conference
• /
• 1997.07a
• /
• pp.352-356
• /
• 1997

This paper present the design of a novel neural-network (NN) based pulse-width modulation (PWM) techniques for a three-level power converter of electric trains along with nonlinear mapping of essential switching patterns and fault tolerance, which are inherent characteristics of NNs. Considering the importance of safety, power factor and harmonics of electric train power converters, two-level type and three-level type of power converters using NNs are precisely investigated and compared in computer simulation. A computer simulation shows that a new current control scheme provides an improved performance over a fixed-band hysteresis current control in many aspects.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.14 no.12
• /
• pp.1287-1294
• /
• 2004

Hub bearings not only sustain the body of a cat, but permit wheels to rotate freely. Excessive radial or axial load and many other reasons can cause defects to be created and grown in each component. Therefore, nitration and noise from unwanted defects in outer-race, inner-race or ball elements of a Hub bearing are what we want to detect as early as possible. How early we can detect the faults has to do with how the detection algorithm finds the fault information from measured signal. Fortunately, the bearing signal has Periodic impulse train. This information allows us to find the faults regardless how much noise contaminates the signal. This paper shows the basic signal processing idea and experimental results that demonstrate how good the method is.

• Proceedings of the KSR Conference
• /
• 2007.05a
• /
• pp.1729-1734
• /
• 2007

Recently, as the feasibility study shows that trans-Korea railway and trans-continental railway are advantageous, interest in high-speed railway system is increasing. Because railway vehicle is environment-friendly and safe compared with airplane and ship, its market-sharing increases gradually. We developed a measurement system for on-line test and evaluation of performances of KHST. The measurement system is composed of software part and hardware part. Perfect interface between multi-users is possible. Nowadays, position data inputs to pulse signal from wheel. Perfect position measurement was limited to slip and slide of vehicle. This measurement makes up for the weak points, Position Detection System using GPS develops. By using the system, Korean High Speed Train is capable of accurate fault position detection.