• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 1996.10a
• /
• pp.37-42
• /
• 1996

기계의 진동.소음은 환경요인과 함께, 기계의 정밀도 향상 및 고장진단과 관련, 기계공학의 중요분야이다. 특히, 전동기(motor)는 많은 기계의 동력원으로서 진동.소음의 1차적 원인을 제공하고 있어, 자체적으로도 저감되어야 함은 물론이며, 감속기등 전동기와 연결되어 사용되는 기계의 진동.소음 발생의 직접적인 원인을 제공하므로 전동기의 진동, 소음 특성은 보다 명확히 밝혀져야 한다. 전동기는 내부에 회전자와 고정자가 있는 회전 기계이면서, 전자기력에 의해 구동되는 전기기계이다. 따라서 축정렬불량(misalignment), 불평형(unbalance)등 기계적 요인과 함께, 고정자와 회전자 사이에 존재하는 공극(air-gap)에서 발생하는 전자기적 요인을 해석함으로써 전동기의 진동.소음의 원인을 밝혀볼 수 있다. 각 전동기에 따라 진동.소음의 크기 및 주파수 성분은 달라질 수 밖에 없으며, 특정 부위에 이상이 있는 경우, 전동기의 진동, 소음특성은 크게 달라지므로, 전동기의 진동, 소음에 관한 연구는 실험적 현상을 기초로 해야한다. 이에 본 연구에서는 전동기이 진동.소음을 규명하기 위해, 부하장치를 가진 전동기 구동 실험장치를 구성하여, 전동기 구동에 따른 진동.소음 신호를 획득, 분석함으로써 전동기에서 발생하는 진동,소음의 원인을 추정하였다.

• Annual Conference of KIPS
• /
• 2013.11a
• /
• pp.1657-1658
• /
• 2013

본 연구에서는 기계 시스템의 위험도 진단 및 관리를 위해, 3차원 기반의 진단결과 가시화 기술에 관한 연구를 소개한다. 계층구조와 인과관계에 의한 신뢰성 분석 기술을 3차원 디지털 목업에 적용하여, 요소 부품 및 전체 시스템의 위험도를 직관적으로 파악하고, 이를 유지 보수 과정에 활용할 수 있는 소프트웨어 플랫폼의 설계 및 풍력발전 시스템 적용을 위한 디지털 목업 개발 등을 소개한다.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.490-495
• /
• 2021

표준어와 방언사이에는 위계가 존재하지 않고 열등하지 않다는 사상을 기반으로 방언을 보존하기 위한 다양한 노력들이 이루어지고있다. 또한 동일한 국가내에서 표준어와 방언간의 의사소통이 잘 이루어져야한다. 본 논문은 방언 연구보존과 의사소통의 중요성을 바탕으로 한국어 방언 기계번역 연구를 진행하였다. 대표적인 방언 중 하나인 제주어와 더불어 강원어, 경상어, 전라어, 충청어 기반의 기계번역 연구를 진행하였다. 공개된 AI Hub 데이터를 바탕으로 Transformer기반 copy mechanism을 적용하여 방언 기계번역의 성능을 높이는 모델링 연구를 진행하였으며 모델배포의 효율성을 위하여 Many-to-one기반 universal한 방언 기계번역기를 개발하였고 이를 one-to-one 모델과의 성능비교를 진행하였다. 실험결과 copy mechanism이 방언 기계번역 모델에 매우 효과적인 요소임을 알 수 있었다.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.629-632
• /
• 2021

기계 번역(machine translation)은 자연 언어로 된 텍스트를 다른 언어로 자동 번역 하는 기술로, 최근에는 주로 신경망 기계 번역(Neural Machine Translation) 모델에 대한 연구가 진행되었다. 신경망 기계 번역은 일반적으로 자기회귀(autoregressive) 모델을 이용하며 기계 번역에서 좋은 성능을 보이지만, 병렬화할 수 없어 디코딩 속도가 느린 문제가 있다. 비자기회귀(non-autoregressive) 모델은 단어를 독립적으로 생성하며 병렬 계산이 가능해 자기회귀 모델에 비해 디코딩 속도가 상당히 빠른 장점이 있지만, 멀티모달리티(multimodality) 문제가 발생할 수 있다. 본 논문에서는 단어 정렬(word alignment)을 이용한 비자기회귀 신경망 기계 번역 모델을 제안하고, 제안한 모델을 한국어-영어 기계 번역에 적용하여 단어 정렬 정보가 어순이 다른 언어 간의 번역 성능 개선과 멀티모달리티 문제를 완화하는 데 도움이 됨을 보인다.

• Annual Conference on Human and Language Technology
• /
• 2020.10a
• /
• pp.353-356
• /
• 2020

관용구는 둘 이상의 단어가 결합하여 특정한 뜻을 생성한 어구로 기계번역 시 종종 오역이 발생한다. 이는 관용구가 지닌 함축적인 의미를 정확하게 번역할 수 없는 기계번역의 한계를 드러낸다. 따라서 신경망 기계 번역(Neural Machine Translation)에서 관용구를 효과적으로 학습하려면 관용구에 특화된 번역 쌍 데이터셋과 학습 방법이 필요하다. 본 논문에서는 한-영 관용구 기계번역에 특화된 데이터셋을 이용하여 신경망 기계번역 모델에 관용구를 효과적으로 학습시키기 위해 특정 토큰을 삽입하여 문장에 포함된 관용구의 위치를 나타내는 방법을 제안한다. 실험 결과, 제안한 방법을 이용하여 학습하였을 때 대부분의 신경망 기계 번역 모델에서 관용구 번역 품질의 향상이 있음을 보였다.

• Journal of Wind Energy
• /
• v.3 no.1
• /
• pp.61-67
• /
• 2012

The loading test of wind turbine pitch and yaw bearings have been conducted using special test rig designed for the test of large slewing bearings. Test type was fatigue test that applied fatigue load to each bearing and followed the defined test process. Measurement data during test were rotational torque and raceway temperature, and inspected key components by disassembling the bearing after all test finished. As a results, the raceway temperature during test did not exceed the operational temperature range of lubricant and rotational torque was reduced as the bearing's rotational cycle increased. In the inspection of key components, some plastic deformation and flaking were detected at some raceway sections while other components such as ball, spacer and seal remain indefective conditions.

• 대한공업교육학회지
• /
• v.44 no.2
• /
• pp.86-103
• /
• 2019

The purpose of this study is to provide basic data on the development of education and training programs for training 'mechanical facilities maintenance manager'. To this end, the DACUM technique was used for job analysis and education and training programs were developed through expert consultation meetings. The job analysis was based on the 10-member DACUM Committee to derive the job definition, job model, job description, and task description of the 'mechanical facilities maintenance manager'. The main findings are as follows. First, the 'mechanical facilities maintenance manager' was defined as those who operate, inspect, diagnose, and repair mechanical facilities to provide the best performance and efficient operation management, provide a safe and pleasant environment, and perform energy saving and facility life extension tasks. Second, the duties of the 'mechanical facilities maintenance manager' analyzed in the job model consist of the comprehensive plan for operation of mechanical facilities, energy management of mechanical facilities, operation management of mechanical facilities, maintenance of mechanical facilities, safety environment management of mechanical facilities, and customer support management of mechanical facilities. Considering the nature and content of the duties, 4 to 11 tasks per duty were derived and a total of 33 tasks were presented as job model. Third, the curriculum for the 'mechanical facilities maintenance manager' was set up in two courses: Practice I for Mechanical Facilities Maintenance and Practice II for Mechanical Facilities Maintenance. Considerations and policy suggestions were presented when applying and implementing education and training programs based on the results of the research.

• Particle and aerosol research
• /
• v.16 no.4
• /
• pp.131-140
• /
• 2020

Air pollutants emitted from chimneys of coal-fired power plants are considered to be a major source of fine particulate matter in the atmosphere. In order to manage fine particle in the chimney of a coal-fired power plant, it is necessary to know the concentration of fine particle emitted in real time, but the current system is difficult. In this study, a real-time measurement system for chimney fine particle was developed, and measurements were performed on six coal-fired power plants. Through the measurements, the mass concentration distribution according to the particle size could be secured. All six chimneys showed bimodal distribution, and the count median diameters of each mode were 0.5 and 1.1 ㎛. In addition, it was compared with the gravimetric measurement method, and it was determined that the relative accuracy for PM10 was within 20%, and the value measured using the developed measuring instrument was reliable. Finally, three power plants were continuously measured for one month, and as a result of comparing the concentration of PM10 according to the amount of power generation, it was confirmed that the PM10 discharged from the chimney increased in the form of an exponential function according to the amount of power generation.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.1
• /
• pp.239-244
• /
• 2022

Neural Machine Translation (NMT) is gradually breaking down the boundary between human and machine translation. We look at actual cases of human and machine translation and discuss why machine translation needs a human touch. In this paper, we raise three driving questions: Can humans be replaced by machines?; How human translators can remain successful in a NMT-driven world?; Is it possible to eliminate language barrier in the era of NMT and World Englishes? The answers to these questions are all negative. We suggest that machine translation is a useful tool with rapidity, accuracy, and low cost productivity. However, the machine translation is limited in the areas of culture, borrowing, ambiguity, new words and (national) dialects. The machines cannot imitate the emotional and intellectual abilities of human translators since machines are based on machine learning, while humans are on intuition. The machine translation will be a useful tool that does not cause moral problems when using methods such as back translation and human post-editing. To conclude, we propose the blended approach that machine translation cannot be completed without the touch of human translation.

• Particle and aerosol research
• /
• v.18 no.1
• /
• pp.9-21
• /
• 2022

In this study, a novel antiviral coating method for the air filtration system of subway station was investigated. Using dry aerosol coating process, we developed a high-performance antiviral air filter with spark discharger and carbon brush type ionizer. Silver nanoparticles were produced by a spark discharge generation system with ion injection system and were used as antiviral agents coated onto a medium grade air filter. The pressure drop, filtration efficiency, and antiviral ability of the filter against aerosolized MS2 virus particles as a surrogate of SARS-CoV-2 virus were tested with dust contamination. Dust contamination caused the increase of the filtration efficiency and pressure drop, while the antiviral agents (in this study, silver nanoparticles) coating did not have any significant effect on the filtration efficiency and pressure drop. Using these properties, we suggested a novel method to maximize the antiviral performance of the antiviral air filter that was contaminated by dust particles. Moreover theoretical analysis of antiviral ability with dust contamination and re-coated antiviral agents was carried out using a mathematical model to calculate the time-dependent antiviral effect of the filter under actual conditions of subway station. Our model can be used to apply on antiviral air filtration system of subway station for prevention of pandemic diffusion, and predict the life cycle of an antiviral filter.