http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Kim, RaeHyun,Son, Yowhan,Hwang, Jaehong The Ecological Society of Korea 2004 Journal of Ecology and Environment Vol.27 No.2
Coarse woody debris (CWD, $\ge$ 5 cm in maximum diameter) is an important functional component, especially to nutrient cycling in forest ecosystems. To compare mass and nutrient dynamics of CWD in natural oak forests, a two-year study was conducted at Quercus serrata and Q. variabilis stands in Yangpyeong, Kyonggi Province. Total CWD (snag, stump, log and large branch) and annual decomposition mass (Mg/ha) were 1.9 and 0.4 for the Q. serrata stand and 7.5 and 0.5 for the Q. variabilis stand, respectively. Snags covered 72% of total CWD mass for the Q. variabilis stand and 42% for the Q. serrata stand. Most of CWD was classified into decay class 1 for both stands. CWD N and P concentrations for the Q. variabilis stand significantly increased along decay class and sampling time, except for P concentration in 2002. There were no differences in CWD N concentration for the Q. serrata stand along decay class and sampling time. However, CWD P concentration decreased along sampling time. CWD N and P contents (kg/ha) ranged from 3.5∼4.7 and 0.8∼1.3 for the Q. serrata stand to 22.8∼23.6 and 3.7∼4.7 for the Q. variabilis stand. Nitrogen and P inputs (kg/ha/yr) into mineral soil through the CWD decomposition were 0.7 and 0.3 for the Q. serrata stand and 1.6 and 0.3 for the Q. variabilis stand, respectively. The number of CWD and decay rate were main factors influencing the difference in CWD mass and nutrient dynamics between both stands.
Raehyun KIM,Peter ATTAYEK,Yuli WANG,Kathleen FURTADO,Rita TAMAYO,Christopher SIMS,Nancy ALLBRITTON 한국생물공학회 2021 한국생물공학회 학술대회 Vol.2021 No.10
Healthy human bodies host hundreds of trillion microbes. More than 99% of these microbes colonize the gut. These gut microbes, also called gut microbiota, have huge impacts on the health and disease states of the hosts. An in vitro intestinal model system that enables co-culturing the gut microbes with human cells can be a valuable tool to elucidate their interaction. However, creating a model system that possesses a proper oxygen gradient to meet different oxygen demands of the gut microbes, primarily obligate anaerobic bacteria, and the human cells, has been challenging. Here, we show a simple in vitro human colon model system that creates the physiological oxygen gradient by utilizing the cellular oxygen consumption and modulating oxygen influx. The behaviors of the primary human colon epithelial cells in the oxygen gradient were compared to those cultured in the conventional aerobic condition. Multiple obligate anaerobic gut bacterial strains were successfully co-cultured in the model system. This novel and simple in vitro colon model system may be a valuable tool to study the interaction of gut bacteria and the host cells.
DeepNAP: Deep neural anomaly pre-detection in a semiconductor fab
Kim, Chunggyeom,Lee, Jinhyuk,Kim, Raehyun,Park, Youngbin,Kang, Jaewoo Elsevier science 2018 Information sciences Vol.457 No.-
<P>Anomaly detection in an industrial process is crucial for preventing unexpected economic loss. Among various signals, multivariate time series signals are one of the most difficult signals to analyze for detecting anomalies. Moreover, labels for anomalous signals are often unavailable in many fields. To tackle this problem, we present DeepNAP which is an anomaly pre-detection model based on recurrent neural networks. Without any annotated data, DeepNAP successfully learns to detect anomalies using partial reconstruction. Furthermore, detecting anomalies in advance is essential for preventing catastrophic events. While previous studies focused mainly on capturing anomalies after they have occurred, DeepNAP is able to pre-detect anomalies. We evaluate DeepNAP and other baseline models on a real multivariate dataset generated from a semiconductor manufacturing fab. Compared with other baseline models, DeepNAP achieves the best performance on both the detection and pre-detection of anomalies. (C) 2018 Elsevier Inc. All rights reserved.</P>
김래현(Kim Raehyun),이우진(Lee Woojin),윤형구(Yoon Hyung-Koo),이종섭(Lee Jong-Sub) 대한토목학회 2009 대한토목학회논문집 C Vol.29 No.4
온도 변화에 의하여 로드셀이나 스트레인 게이지의 결과값이 변하기 때문에 로드셀이나 전기저항식 변형율계를 이용하는 기계적 장치는 주변 온도의 변화에 의하여 영향을 받는다. 본 연구에서는 기존의 전기저항식 변형율계 타입의 콘이 가지는 문제점과 한계를 극복하고 관입과 동시에 연속적으로 온도보상이 가능한 직경 1~7㎜의 초소형 광섬유 마이크로콘을 개발하였다. 광섬유는 머리카락 굵기의 작은 직경을 가지고 있어 원하는 크기의 센서를 구성할 수 있고 전기적 신호인 전압을 측정하는 것이 아니라 빛의 파장변화를 감지하여 변형율로 변환하게 되므로 주변 조건에 의한 간섭 영향이 거의 없다. 개발된 콘의 온도보상 효과를 검증하기 위해 외력이 없는 상태에서 큰 주변의 온도를 변화시키는 용도시험을 실시하였다. 주변온도에 따라 측정전압이 변화되는 전기저항식 변형률계 콘과는 달리 광섬유 센서를 적용한 콘은 일정한 값을 유지하는 것으로 나타났으며 관입과 동시에 지중의 온도변화를 연속적으로 확인할 수 있었다. 또한, 다층지반 콘관입시험에서는 관입되는 동안 교란영역이 적고 분해능이 뛰어나 선단지지력의 변화만으로도 다층지반의 경계를 명확하게 탐지할 수 있는 것으로 분석되었다. 본 연구에서는 광섬유 센서를 이용한 초소형 마이크로콘으로 다층지반을 효과적으로 탐지할 수 있으며 지중의 온도 영향이 고려된 순수한 선단저항력을 획득할 수 있음을 보여준다. Mechanical device using the load cell or strain gage sensor can be influenced by tempearute changes because temperature change can cause a shift in the load cell or straing gage output at zero loading. In this paper, micro cone penetrometers with 1~7㎜ in diameter, are developed by using an optical fiber sensor (FBG: Fiber Bragg Grating) to compensate the continous temperature change during cone penetration test. Note the temperature compensated method using optical fiber sensor which has hair-size in diameter, and is not affected by environmental conditions because the measured data is the wavelength shifting of the light instead of the intensity of the electric voltage. Temperature effect test shows that the output voltage of strain gage changes and increases with an increase in the temperature. A developed FBG cone penetrometer, however, achieves excellent temperature compensation during penetration, and produces continuous change of underground temperature. In addition, the temperature compensated FBG cone shows the excellent sensitivity and detects the interface of the layered soils with higher resolution. This study demonstrates that the fiber optic sensor renders the possibility of the ultra small size cone and the new fiber optic cone may produce more reliable temperature compensated tip resistance.
AVX-512를 활용한 인텔 프로세서에서 행렬 곱셈 연산의 오토튜닝 방법
이영하(Yeongha Lee),김래현(Raehyun Kim),최재영(Jaeyoung Choi) 한국정보과학회 2018 정보과학회 컴퓨팅의 실제 논문지 Vol.24 No.12
BLAS는 선형 대수 및 행렬 연산을 수행할 때 널리 사용되는 라이브러리이다. 특히 블럭 연산을 지원하는 level-3 BLAS는 캐시에 올라오는 데이터의 재사용성을 높여 성능을 최대로 얻을 수 있도록 한다. 이를 위해서는 프로세서의 캐시에 맞는 행렬의 블록크기, 루프 언롤링(loop unrolling) 횟수, 프리패칭(prefetching) 거리 등이 최적화 되는 설정값을 찾아내야 한다. 이를 컴퓨터가 자동으로 실행하면서 최적화하는 과정을 오토튜닝이라고 한다. Intel Knights Landing 등 Intel AVX-512를 사용하는 최신의 인텔 프로세서들은 BLAS 최적화를 위한 오토튜닝 방법이 거의 연구되어 있지 않으므로, 본 연구에서는 Intel AVX-512를 사용하는 인텔의 아키텍처들에서 level-3 BLAS의 성능을 최대화하기 위한 오토튜닝 방법을 보인다. BLAS is a widely used library for performing linear algebra and matrix operations. Level-3 BLAS, in particular, supports block operations to maximize performance by increasing the reusability of data in the cache. For a block operation, it is necessary to find optimal values that optimize the block size of the matrix matched to the processor`s cache, the number of loop unrolling times, and the prefetching distance. Autotuning means that the computer runs and automatically optimizes its performance. The appropriate autotuning processes for BLAS routines have not studied on the recent Intel processors with AVX-512 such as Intel Knights Landing and Intel Scalable processors. In this study, we have demonstrated an autotuning method to maximize the performance of level-3 BLAS in Intel architectures with Intel AVX-512.