• 대한전기학회논문지:전력기술부문A
• /
• 제49권8호
• /
• pp.390-398
• /
• 2000

Leakage impulse current of the contaminated insulators by using experiment data were studied. The impulse current in phase-time relationship was analyzed on line post insulators. Also, frequency components and crest factor of the leakage current were investigated to provide a scheme for an early detection of insulator incipient failure. The study shows that the phase-time characteristic is non-stationary and random and, non-harmonic component and crest factor can be promising parameters for detecting insulator leakage currents.

• 조명전기설비학회논문지
• /
• 제22권2호
• /
• pp.71-79
• /
• 2008

철도 고압배전선로의 경우 궤도를 따라 양방향으로 선로연변에 통신 및 신호설비와 병행하여 가공 또는 지중선로로 설치되어 있다. 가공선로의 경우에는 대기중에 노출되어 있어 뇌격, 폭풍우, 염해 등 자연현상으로 인한 고장발생이 다수 발생하고 있으며, 이에 따른 보호장치의 오 부동작이 빈번하게 발생하고 있다. 철도 고압배전선로에서 발생하는 사고 중 가장 많은 것은 1선 지락이지만 이밖에 선간 단락, 심할 경우에는 3선 지락(단락)으로까지 진전되는 사고가 있을 뿐만 아니라 단선 사고까지 발생하는 경우도 있다. 따라서 사고를 방지하기 위해서는 보다 상세한 점검보수가 필요하며, 고장발생시 조기발견과 신속한 고장처리는 철도안전수송에 중요하다. 본 논문에서는 철도 고압배전계통의 주류를 이루게 될 22.9[kV] 직접접지 계통을 대상으로 고장 발생시 고장 위치를 신속하게 표정할 수 있는 고장점 표정 알고리즘 개발을 위해 22.9[kV] 고압배전계통을 모델링 하여 특성해석과 고장해석을 수행하였고, 정확한 고장점 표정이 가능한 반복계산법을 이용한 알고리즘을 제시하였으며, 사례연구를 통해 성능을 입증하였다.

• 한국소음진동공학회논문집
• /
• 제25권8호
• /
• pp.553-558
• /
• 2015

This paper aims at making a dedicated vibration monitoring and diagnosis framework for 3MW WTG(wind turbine generator). Within the scope of the research, vibration data of WTG drive train are used and WTG operating conditions are involved for dividing the vibration data class which included transient and steady state vibration signals. We separate two health detections which are CHD(continuous health detection) and EHD(event health detection). CHD has function of early detection and continuous monitoring. EHD makes the use of finding vibration values of fault components effectively by spectrum matrix subsystem. We proposed framework and showed application for 3MW WTG in a practical point of view.

• 전기전자학회논문지
• /
• 제26권2호
• /
• pp.319-328
• /
• 2022

베어링은 기계가 작동할때 중요한 역할을 한다. 때문에, 베어링에 결함이 발생하면 기계전체의 치명적인 결함을 발생시킨다. 그러므로 베어링 결함은 조기에 발견되어야한다. 본 논문에서는 연속 웨이블릿 변환과 Switchable 정규화를 기반으로 한 합성곱 신경망(SN-CNN)을 이용한 방법을 베어링 결함 감지 모델에 대해 설명한다. 모델의 정확도는 Case Western Reserve University(CWRU) 베어링 데이터 집합을 사용하여 측정되었다. 또한 배치 정규화(BN, Batch Normalization)[1] 방법과 스펙트로그램 이미지가 모델 성능의 비교를 위해 사용되었다.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2007년도 추계학술대회 논문집
• /
• pp.903-912
• /
• 2007

Faults of rotating parts of a train normally generate unexpected frequency band or impulsive sound[1] which has a period when it moves with a constant speed. The former can be detected by the moving frame acoustic holography method, which visualizes sound field that is generated by a moving and emitting pure tone or band limited noise source. We have attempted to apply the method to the latter case: the periodic impulsive sound which generate different signal compared with what can be measured by the band limited noise. The signal to noise ratio which determines the success of early fault detection must also be studied with the impulsive and moving signal. This research shows how the problems related with these issues can be resolved. The main idea is that periodic impulsive signal can be expressed by infinite set of discrete pure tones. This enables us to obtain lots of holograms that visualize periodic impulsive sound field including noise by using the moving frame acoustic holography method. Therefore holograms can be averaged to improve the signal to noise ratio until having reliable information that exhibits where the impulsive sources are. Theory and experiment by using the miniature vehicle are described [Work supported by BK21 & KRRI].

• Journal of Electrical Engineering and Technology
• /
• 제13권1호
• /
• pp.1-10
• /
• 2018

Wide-area monitoring of tree-related high impedance fault (THIF) efficiently contributes to increase reliability of large-scaled network, since the failure to early location of them may results in critical lines tripping and consequently large blackouts. In the first place, this wide-area monitoring of THIF requires managing the placement of sensors across large power grid network according to THIF detection objective. For this purpose, current paper presents a framework in which sensors are distributed according to a predetermined risk map. The proposed risk map determines the possibility of THIF occurrence on every branch in a power network, based on electrical conductivity of trees and their positions to power lines which extracted from spectral data. The obtained possibility value can be considered as a weight coefficient assigned to each branch in sensor placement problem. The next step after sensors deployment is to on-line monitor based on moving data window. In this on-line process, the received data window is evaluated for obtaining a correlation between low frequency and high frequency components of signal. If obtained correlation follows a specified pattern, received signal is considered as a THIF. Thereafter, if several faulted section candidates are found by deployed sensors, the most likely location is chosen from the list of candidates based on predetermined THIF risk map.

• Smart Structures and Systems
• /
• 제30권6호
• /
• pp.613-626
• /
• 2022

Guaranteeing the quality and integrity of structural health monitoring (SHM) data is very important for an effective assessment of structural condition. However, sensory system may malfunction due to sensor fault or harsh operational environment, resulting in multiple types of data anomaly existing in the measured data. Efficiently and automatically identifying anomalies from the vast amounts of measured data is significant for assessing the structural conditions and early warning for structural failure in SHM. The major challenges of current automated data anomaly detection methods are the imbalance of dataset categories. In terms of the feature of actual anomalous data, this paper proposes a data anomaly detection method based on data-level and deep learning technique for SHM of civil engineering structures. The proposed method consists of a data balancing phase to prepare a comprehensive training dataset based on data-level technique, and an anomaly detection phase based on a sophisticatedly designed network. The advanced densely connected convolutional network (DenseNet) and Transformer encoder are embedded in the specific network to facilitate extraction of both detail and global features of response data, and to establish the mapping between the highest level of abstractive features and data anomaly class. Numerical studies on a steel frame model are conducted to evaluate the performance and noise immunity of using the proposed network for data anomaly detection. The applicability of the proposed method for data anomaly classification is validated with the measured data of a practical supertall structure. The proposed method presents a remarkable performance on data anomaly detection, which reaches a 95.7% overall accuracy with practical engineering structural monitoring data, which demonstrates the effectiveness of data balancing and the robust classification capability of the proposed network.

• Journal of Electrical Engineering and Technology
• /
• 제4권2호
• /
• pp.219-228
• /
• 2009

The growing application and the numerous qualities of induction motors (1M) in industrial processes that require high security and reliability levels has led to the development of multiple methods for early fault detection. However, various faults can occur, such as stator short-circuits and rotor failures. Traditionally the diagnosis machine is done through a sinusoidal power supply, in the present paper we study experimentally the effects of the rotor failures, such as broken rotor bars in function of the ac supplying, the load and show the impact of the converter from diagnosis of the machine. The technique diagnosis used is based on the spectral analysis of stator currents or stator voltages respectively according to the types of induction motor ac supplying. So, four different ac supplying are considered: ${\odot}$ the IM is directly by the balanced three-phase network voltage source, ${\odot}$ the IM is fed by a sinusoidal current source given the controlled by hysteresis, ${\odot}$ the IM is fed (in open loop) by a scalar control imposing through ratio V/f=constant, ${\odot}$ the IM is controlled through a vector control using space vector pulse width modulation (SVPWM) technique inverter with an outer speed loop.

• 한국음향학회지
• /
• 제30권8호
• /
• pp.446-460
• /
• 2011

최근 항공 산업, 자동차 산업 등의 산업 현장에서 유도 전동기의 사용이 증대되고 있으며, 유도 전동기는 산업 현장에서 중요한 역할을 하고 있다. 따라서 유도 전동기의 고장으로 인한 피해를 최소화하기 위해 유도 전동기의 고장 검출 및 분류 시스템의 개발이 중요한 문제로 대두되고 있다. 이와 같은 이유로 본 논문에서는 유도 전동기의 고장을 조기에 검출하고 진단하기 위해 에너지 (short-time energy)와 특이치 분해와 이산 코사인 변환과 특이치 분해를 이용한 특징 벡터 추출 방법을 제안하였고, 이를 역 전파 신경 회로망과 다층 서포트 벡터 머신의 입력으로 이용하여 유도 전동기의 고장을 유형별로 분류하였다. 하지만 본 논문에서는 역 전파 신경 회로망과 다층 서포트 벡터 머신을 분류기로 사용함에 있어 역 전파 신경 회로망은 신경망을 구성하는 입력 뉴런 수, 은닉 뉴런 수, 학습 알고리즘에 의해 분류 성능이 달라지며, 다층 서포트 벡터 머신은 커널 함수로 사용한 가우시안 방사 기저 함수의 표준 편차 값에 따라 분류 성능이 달라지는 점을 고려하여 여러 가지 조건하에서의 실험을 통해 높은 분류 성능을 보이는 설정 방법을 제시하였다.

• 품질경영학회지
• /
• 제47권1호
• /
• pp.75-86
• /
• 2019

Purpose: It is essential for the steel industry to produce steel products without unexpected downtime to reduce costs and produce high quality products. A hot strip rolling mill consists of many mechanical and electrical units. In condition monitoring and diagnosis, various units could fail for unknown reasons. Methods: In this study, we propose an effective method to detect units with abnormal status early to minimize system downtime. The early warning problem with various units was first defined. An autoencoder was modeled to detect abnormal states. An application of the proposed method was also implemented in a simulated field-data analysis. Results: We can compare images of original data and reconstructed images, as well as visually identify differences between original and reconstruction images. We confirmed that normal and abnormal states can be distinguished by reconstruction error of autoencoder. Experimental results show the possibility of prediction due to the increase of reconstruction error from just before equipment failure. Conclusion: In this paper, hot strip roughing mill monitoring method using autoencoder is proposed and experiments are performed to study the benefit of the autoencoder.