http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Design of a Fault-Detection System for FDM-type 3D Printer using Temporal Convolutional Network
Danielle Jaye S. Agron,Gabriel Avelino R. Sampedro,Gabriel Amaizu,Jae-Min Lee,Dong-Seong Kim 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.6
In the process of additive manufacturing, the devices used to print usually encounter errors and problems that are not easily detected by the device operator. Undetected errors can cause serious damage to the 3D printer and leads to the output being counted as reject, thus leading to both loss in time and resources. The research focuses on the development of a device to monitor the process of 3D printing. The design applies temporal convolutional networks (TCN) to train the device to identify whether certain measurements of the 3D printer will lead to errors in output. The prototype serves as an attachment to the 3D printer and displays measurements and if they are within the safe values.
Danielle Jaye S. Agron,Jae-Min Lee,Dong-Seong Kim 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.2
A printing technique called fused deposition modelling (FDM) in additive manufacturing make use of Polylactic acid (PLA) material by melting in nozzle that when stacked layer-by-layer it forms a rigid final product. However, due to varying nozzle melting temperature the risk of thermal degradation escalates. In order to avoid that, this paper proposed a scheme to predict the anomaly during the printing process using a shallow recurrent neural network; a shallow recurrent neural network (O-LSTM) framework. A time series thermal estimation is provided to forecast the occurrence of error with a 90.9% accuracy.
Development of ultracapacitor management system controller for solar-powered streetlamp
Danielle Jaye S. Agron(다니),Henar Mike O. Canilang(헤나르),Angela Caliwag(안젤라),Wansu Lim(임완수) 한국통신학회 2022 한국통신학회 학술대회논문집 Vol.2022 No.2
In this paper, an ultra-capacitor management system (UCMS) controller for solar-powered streetlamp is developed. To increase the efficiency and deployment life of the ultra-capacitor, an active balancing scheme has been applied for the charging and discharging phase. Through the balancing approach, an active charge and discharge protection control capability is integrated for this application realizing a fault tolerant system. The development of hardware prototype and initial results are presented on this paper.
저비용 우주 발사체 개발 동향 및 이를 위한 차세대 연료에 대한 고찰
배진현(Jinhyun Bae),구자예(Jaye Koo),윤영빈(Youngbin Yoon) 한국항공우주학회 2017 韓國航空宇宙學會誌 Vol.45 No.10
인공위성의 경량화 및 소형화로 인하여 대형발사체보다는 발사 비용이 저렴한 저비용 발사체에 대한 관심이 증가되고 있다. 저비용 발사체의 비용 절감 중 가장 대표적인 방식이 발사체의 재사용이다. 저비용 발사체를 개발하고 있는 대부분의 기업들 역시 발사체 재사용 방식을 채택하고 있다. 이러한 재사용 목적과 더불어 친환경 우주 발사체에 대한 요구가 증가되면서 저비용 발사체에 사용되는 연료의 선택 역시 매우 중요해졌다. 친환경적이면서 발사체의 재사용이 가능하게 하는 연료 중 에너지 밀도 등 다른 요인을 고려했을 때 가장 적합한 것이 메탄이며, 메탄에 수소를 첨가하여 에너지 밀도를 높게 만든 HCNG(hydrogen-enriched compressed natural gas) 역시 적합하다고 판단되었다. 본 연구는 한국형 발사체 개발 이후 국내 우주 개발 방향 설정의 참고자료로써 전 세계 저비용 발사체 동향 및 로켓 연료의 특성에 대해 고찰하였다. Due to the weight reduction and miniaturization of satellites, there is a growing interest in low-cost launch vehicles, which are cheaper to launch than larger launch vehicles. One of the most cost-effective ways to reduce the cost of launch vehicles is the reuse of vehicles. Most companies that are developing low cost launch vehicles are also adopting a vehicles reuse approach. Along with this reuse purpose, the demand for environmentally friendly space launch vehicles has increased, so the choice of fuel used for low cost launch vehicles has also become very important. Methane and hydrogen-enriched compressed natural gas (HCNG), which makes more energy-efficient by adding hydrogen to methane, are considered to be the most suitable when considering other factors such as energy density among the fuels that are eco-friendly and capable of reusing the launch vehicles. This study investigated the trends of low-cost launch vehicle and rocket fuel in the world as a reference for setting up domestic space development after the development of Korea Space Launch Vehicle-II.
김원일,이재명,강종표 한국산업안전학회 1997 한국안전학회지 Vol.12 No.4
From the experimental study of wire-cut Electric Discharge Machining for alloyed steel and tungsten carbide, the characteristics such as hand drum form has been observed and evaluated for various conditions. Hand drum form can be improved when gap voltage and spark cycle become smaller, their thickness become thinner, wire tension become larger and number of cutting is done so many times. When wire-cut 60㎜ thickness tungsten carbide in normal condition, Hand drum form becomes larger due to the low conductivity inducing cobalt composite rising by electrolysis.
축소형 연소기에서 임계 조건에 따른 화염구조 가시화 시험
송우석(Wooseok Song),구자예(Jaye Koo) 한국추진공학회 2019 한국추진공학회 학술대회논문집 Vol.2019 No.5
고성능 액체로켓엔진 개발을 위해서는 고온, 고압의 연소가 필수적이다. 본 논문에서는 축소형 연소기에서 임계 조건에 따른 화염구조를 가시화하는 것이 목표이다. 전단 동축 분사기를 이용하여 산화제는 기체산소를 사용하였고 연료는 액체 케로신을 사용하였다. 화염구조를 촬영하기 위해 탄화수소계 연료 연소에서 생성되는 CH* 화학발광 성분을 밴드패스필터 및 고속카메라를 이용하여 가시화하였다. 연소불안정 정도를 계산하여 아임계/초임계 연소조건에서 안정된 화염을 확인하였다. 안정된 연소조건에서 화염두께는 아임계 연소조건보다 초임계 연소조건에서 작게 측정되었다.. The high temperature and pressure in the combustion chamber is essential to develop the liquid rocket engine for a high performance. The objective of this study is to visualize the flame structure under subcritical and supercritical conditions using a subsclae combustor. The gaseous oxygen and liquid kerosene were used for propellants with the shear coaxial injector. In order to detect the flame structure, CH* chemiluminescience images, which is one of representative species during the combustion process, was recorded using band-pass filter and high-speed camera. A stable flame was confirmed by calculating the combustion instability intensity under subcritical and supercritical conditions. The flame thickness in the case of supercritical condition was thiner than in the case of subcritical condition.
Automated Fall Detection on Smart Factory based on Deep Learning Approach
Syifa Maliah Rachmawati,Danielle Jaye S. Agron,Dong-Seong Kim,Jae-Min Lee 한국통신학회 2021 한국통신학회 학술대회논문집 Vol.2021 No.6
The emergence of the smart environment and the Internet of Things paradigms with the increasing number of cameras in daily life, forms an optimal context for vision-based systems. This paper proposes a model to detect human falling by using deep learning algorithm for vision-based system. To improve fall-detection accuracy, a deep learning technique such as Convolutional Neural Network (CNN) combined with data augmentation and dropout layer to avoid over-fitting is proposed. It compared with existing convolutional-based architecture such as AlexNet, SqueezeNet, GoogleNet, and ResNet-18. The per-formance of the proposed algorithm is verified by using UR Fall Detection data set. The simulation result showed that the proposed algorithm achieves an accuracy 96.88% with validation loss 0.0638.