• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2008년도 추계학술대회 논문집
• /
• pp.235-240
• /
• 2008

Since urban railway transit is one of the most essential transportation systems, its power facilities must ensure high reliability and safety. Currently, urban railway operating organizations perform TBM (Time Based Maintenance) on power facilities. However, in order to improve management efficiency and system safety, CBM (Condition Based Maintenance) is preferred. Among various power facilities, mold transformers has been chosen as the object of study since it is widely used for the purpose of minimizing volume and weight, and due to safety against fire. In this paper, various transformer failure cases due to electric, thermal, mechanical and environmental factors have been collected and analyzed. In addition, investigation on national and international condition based maintenance cases and the characteristics of sensors widely used for transformer monitoring has been performed to suggest the optimal condition based maintenance technique for urban railway systems.

• Journal of Information Processing Systems
• /
• 제18권2호
• /
• pp.282-291
• /
• 2022

This research applies a pre-trained bidirectional encoder representations from transformers (BERT) handwriting recognition model to predict foreign Korean-language learners' writing scores. A corpus of 586 answers to midterm and final exams written by foreign learners at the Intermediate 1 level was acquired and used for pre-training, resulting in consistent performance, even with small datasets. The test data were pre-processed and fine-tuned, and the results were calculated in the form of a score prediction. The difference between the prediction and actual score was then calculated. An accuracy of 95.8% was demonstrated, indicating that the prediction results were strong overall; hence, the tool is suitable for the automatic scoring of Korean written test answers, including grammatical errors, written by foreigners. These results are particularly meaningful in that the data included written language text produced by foreign learners, not native speakers.

• 대한임베디드공학회논문지
• /
• 제17권6호
• /
• pp.319-327
• /
• 2022

Transformers are the most famous deep learning models that has achieved great success in natural language processing and also showed good performance on computer vision. In this survey, we categorized transformer-based models for computer vision, particularly object detection tasks and perform comprehensive comparative experiments to understand the characteristics of each model. Next, we evaluated the models subdivided into standard transformer, with key point attention, and adding attention with coordinates by performance comparison in terms of object detection accuracy and real-time performance. For performance comparison, we used two metrics: frame per second (FPS) and mean average precision (mAP). Finally, we confirmed the trends and relationships related to the detection and real-time performance of objects in several transformer models using various experiments.

• Journal of Information Processing Systems
• /
• 제19권1호
• /
• pp.80-88
• /
• 2023

RGB-thermal (RGB-T) tracking using unmanned aerial vehicles (UAVs) involves challenges with regards to the similarity of objects, occlusion, fast motion, and motion blur, among other issues. In this study, we propose dynamic tracking aggregation (DTA) as a unified framework to perform object detection and data association. The proposed approach obtains fused features based a transformer model and an L1-norm strategy. To link the current frame with recent information, a dynamically updated embedding called dynamic tracking identification (DTID) is used to model the iterative tracking process. For object association, we designed a long short-term tracking aggregation module for dynamic feature propagation to match spatial and temporal embeddings. DTA achieved a highly competitive performance in an experimental evaluation on public benchmark datasets.

• 방재기술
• /
• 통권12호
• /
• pp.36-46
• /
• 1992

This report is on how to extend the life of transformer and reduce losses resulted from fire and faiure of transformer through the proper maintenance and fire protection of Oil-insulated transformer

• 전력전자학회지
• /
• 제22권2호
• /
• pp.38-45
• /
• 2017

• Korean Journal of Radiology
• /
• 제25권1호
• /
• pp.113-115
• /
• 2024

• 전기학회논문지
• /
• 제63권2호
• /
• pp.266-271
• /
• 2014

The Electromagnetic Interference(EMI) generating from corona discharge of transformer area can interference the digital Instrument and Control(I&C) systems located nearby transformers. When the potential gradient of the electric field around the conductor is high enough to form a conductive region but not high enough to cause electrical breakdown to nearby objects, the EMI of corona discharge emits with the conducted and radiated noise and it interferences the signals of the I&C systems. Since digital I&C systems have an efficiency and competitive price, the analog I&C systems have been upgraded and displaced with the digital I&C systems but which have less EMI Immunity. There was no assessment to I&C systems by EMI generating corona discharge nearby transformers. When the safety-related I&C systems are installed in plants, the verification of equipment EMI should be done not in site-specific test but in test facilities. There are the need to do the site-specific EMI evaluation of corona discharge nearby transformers. This paper assesses the margin between plant emission limits and the highest composite plant emission of corona. When the non safety-related I&C systems are placed in transformer area, it suggests the appropriate radiated susceptibility level to EMI of corona discharge.

• 조명전기설비학회논문지
• /
• 제21권1호
• /
• pp.104-110
• /
• 2007

본 논문은 권선내부에 온도계가 설치된 지상변압기를 제작하여 부하운전에 따른 배전용 변압기의 온도상승 특성과 절연유의 유전특성 변화를 관찰하였다. 관찰결과 배전용변압기의 경우 과열점은 세 권선 중 두개의 권선이 접하고 권선의 약 2/3 높이에 위치하고 있으며, 과열점과 절연유 상부 간에 약 $17[^{\circ}C]$의 일정한 온도차를 확인할 수 있었다. 또한 절연유의 정전용량특성은 온도와 선형적인 관계로 감소하는 특성을 보이며, 그 기울기는 열화율과 무관하게 -0.0106의 일정한 값을 관찰할 수 있었다. 이 결과는 수명손실의 정확한 계산과 열화진단을 위한 기구로 활용할 수 있을 것이다.

• 전력전자학회논문지
• /
• 제8권2호
• /
• pp.99-106
• /
• 2003

본 논문에서는 출력 전압 파형을 개선하고 고조파 성분을 저감시키기 위한 한 방법으로 변압기 2차측을 직렬로 결합시킨 형태를 가지는 새로운 멀티 레벨 인버터를 제안한다. 제안된 11 레벨 PWM 인버터는 출력 전압의 각 레벨을 형성하기 위한 두 개의 level 인버터 모듈과 PWM 스위칭 동작을 위한 PWM 인버터 모듈, 그리고 3대의 변압기로 구성된다. 적절한 변압기의 권선비를 이용하여 직류 전원에 대한 정수비의 연속적인 출력 전압 레벨을 형성하였으며, 변압기의 직렬 결합으로 출력단의 필터용 인덕허가 필요 없는 장점을 가진다. 제안하는 PWM 인버터의 타당성을 입증하기 위해 24 [V] 직류 전원에서 220 [V] 교류 전원을 발생시킬 수 있는 인버터의 시작품을 제작하고 실험을 행하였으며, 기존의 멀티 레벨 방식을 이용한 11 레벨 PWM 인버터와 제안된 11 레벨 PWM 인버터의 비교를 통해 제안하는 인버터의 타당성을 검증하였다.