• 한국정보과학회논문지:소프트웨어및응용
• /
• 제34권7호
• /
• pp.655-666
• /
• 2007

RacerPro, Pellet 등 지금까지의 전혈적인 추론 시스템들은 주로 Tableaux Algorithm 기반의 추론 시스템으로 Tableaux Algorithm의 특성상 대용량 ABox 추론에서 문제점을 나타낸다. 이를 해결하기 위한 연구로 Tableaux Algorithm 기반에 DBMS를 함께 사용한 영국 Manchester 대학의 Instance Store와 Disjunctive Datalog Approach를 사용한 독일 Karlsruhe 대학의 KAON2가 있다. 현재 추론 시스템들에 대한 벤치마크 실험은 대부분 Tableaux Algorithm 기반의 TBox 추론 위주이며 ABox 추론에 대한 평가는 거의 진행되지 않았다. 특히 최근 이슈로 부각된 (대용량 ABox 추론을 위한 추론 시스템)의 특성별 벤치마크 실험은 거의 보고되지 않았다. 이에 본 논문에서는 각 추론엔진들의 이론적 배경을 근간으로 전형적 추론엔진들과 최근 이슈에 따른 대용량 ABox론 위한 추론엔진들을 상호 비교를 통해 살펴보며 특히, 대용량 ABox 처리론 위한 추론엔진인 Manchester 대학의 Instance Store와 Karlsruhe 대학의 KAON2를 LUBM을 통하여 분석 평가함으로 사용자의 요구에 따른 대용량 ABox 추론엔진을 제시한다. 평가방법에서는 LUBM(Lehigh University BenchMark)에 대한 소개와 이를 이용한 벤치마크 실험 방법 및 평가 시스템에 대하여 소개한다. 본 논문은 결론을 통해 실험 결과와 각 추론엔진의 사용 Algorithm 특성을 기초로 다양한 환경에서의 대용량 ABox 처리에 적합한 추론엔진을 제시한다.

• 물과 미래
• /
• 제23권1호
• /
• pp.6-20
• /
• 1990

산업이 고도로 발전하면서 많은 나라들은 인간의 생활 환경을 인간과 자연이 조화로운 삶을 영위할 수 있는 공원과 같은 생활공간으로 개발하기 위하여 많은 부존자원을 이용하고 있다. 이들 자원중에서 수자원은 쾌적한 생활공간을 조성하기 위한 가장 중요한 것으로서, 효율적인 수자원 이용을 위하여 물의 사용량 결정, 사용량 평가 등에 관한 사항들을 정량화하기 위하여 수문모형을 사용하고 있다. 본 논문은 독일의 Karlsruhe 대학의 Institute for Hydrology and Water Resourced Planning(IHW)에서 수행된 연구에 기초를 두고 있는 것으로서 장래의 강우-유출 모형개발을 고찰하였다.

• 대한화학회지
• /
• 제24권5호
• /
• pp.371-379
• /
• 1980

테트라히드로퓨란(THF)-물 혼합용매에서 용매화된 전자($e^-_s$)를 생성시켜 광흡수스펙트럼의 성질을 연구하였다. 본 연구에서 $e^-_s$는 nsec Q 스위치 Nd(YAG) 레이저 에너지 (265nm)를 이용한 광분해법으로 생성시켰다. 이 방법은 실험부분에서 도면으로 간단하게 기술되어 있다. 측정한 $e^-_s$의 최대흡수 파장은 THF의 함량과 온도에 따라 장파장쪽으로 이동하였고, 흡수스펙트럼의 폭 역시 THF에 따라 증가하였다. $e^-_s$의 반감기 또한 THF의 몰%에 따라 증가하였다. 혼합용매에서 최대 흡수파장은 순수한 물이나 THF에서 방사선 분해법을 측정한 흡수파장의 중간에서 모두 발견되었다. 이러한 실험 결과로부터 $e^-_s$의 체적과 최대흡수에너지 관계를 논의할 수 있었다.

• Nuclear Engineering and Technology
• /
• 제42권6호
• /
• pp.656-661
• /
• 2010

Lead-alloys are very attractive nuclear coolants due to their thermo-hydraulic, chemical, and neutronic properties. By utilizing the HELIOS (Heavy Eutectic liquid metal Loop for Integral test of Operability and Safety of PEACER$^2$) facility, a thermal hydraulic benchmarking study has been conducted for the prediction of pressure loss in lead-alloy cooled advanced nuclear energy systems (LACANES). The loop has several complex components that cannot be readily characterized with available pressure loss correlations. Among these components is the core, composed of a vessel, a barrel, heaters separated by complex spacers, and the plenum. Due to the complex shape of the core, its pressure loss is comparable to that of the rest of the loop. Detailed CFD simulations employing different CFD codes are used to determine the pressure loss, and it is found that the spacers contribute to nearly 90 percent of the total pressure loss. In the system codes, spacers are usually accounted for; however, due to the lack of correlations for the exact spacer geometry, the accuracy of models relies strongly on assumptions used for modeling spacers. CFD can be used to determine an appropriate correlation. However, application of CFD also requires careful choice of turbulence models and numerical meshes, which are selected based on extensive experience with liquid metal flow simulations for the KALLA lab. In this paper consistent results of CFX and Star-CD are obtained and compared to measured data. Measured data of the pressure loss of the core are obtained with a differential pressure transducer located between the core inlet and outlet at a flow rate of 13.57kg/s.

• The Journal of the Acoustical Society of Korea
• /
• 제21권1E호
• /
• pp.3-11
• /
• 2002

Broadcast news transcription is one of the hardest tasks in speech recognition because broadcast speech signals have much variability in speech quality, channel and background conditions. We developed a Korean broadcast news speech recognizer. We used a morpheme-based dictionary and a language model to reduce the out-of·vocabulary (OOV) rate. We concatenated the original morpheme pairs of short length or high frequency in order to reduce insertion and deletion errors due to short morphemes. We used a lexicon with multiple pronunciations to reflect inter-morpheme pronunciation variations without severe modification of the search tree. By using the merged morpheme as recognition units, we achieved the OOV rate of 1.7% comparable to European languages with 64k vocabulary. We implemented a hidden Markov model-based recognizer with vocal tract length normalization and online speaker adaptation by maximum likelihood linear regression. Experimental results showed that the recognizer yielded 21.8% morpheme error rate for anchor speech and 31.6% for mostly noisy reporter speech.

• 마이크로전자및패키징학회지
• /
• 제20권4호
• /
• pp.1-6
• /
• 2013

Recently, various types of flexible devices such as flexible displays, batteries, e-skins and solar cell panels have been reported. Most of the researches focus on the development of high performance flexible device. However, to realize these flexible devices, the long-term reliability should be guaranteed during the repeated deformations of flexible devices because the direct mechanical stress would be applied on the electronic devices unlike the rigid Si-based devices. Among various materials consisting electronics devices, metal electrode is one of the weakest parts against mechanical deformation because the mechanical and electrical properties of metal films degrade gradually due to fatigue damage during repeated deformations. This article reviews the researches of fatigue behavior of thin metal film, and introduces the methods to enhance the reliability of metal electrode for flexible device.

• Journal of the Optical Society of Korea
• /
• 제11권1호
• /
• pp.34-39
• /
• 2007

Laser-induced breakdown of colloidal suspensions, such as polystyrene, $ZrO_2$, and $SiO_2$ particles in diameters of 100-400 nm in water is investigated by nanosecond flash-pumped Nd:YAG laser pulses operating at a wavelength ${\lambda}$= 532 nm. The breakdown threshold intensity is examined in terms of breakdown probability as a function of laser pulse energy. The threshold intensity for $SiO_2$ particles ($1.27{\times}10^{11}\;W/cm^2$) with a size of 100 nm is higher than those for polystyrene and $ZrO_2$ particles with the same size, namely $5.7{\times}10^{10}$ and $5.5{\times}10^{10}\;W/cm^2$, respectively. Results indicate that the absorption of five photons is required to induce ionization of $SiO_2$ particles, whereas the other particles necessitate four-photon absorption. These breakdown thresholds are compared with those measured by nanosecond pulses from a diode-pumped Nd:YAG laser having a different focusing geometry.

• 한국세라믹학회지
• /
• 제49권1호
• /
• pp.19-28
• /
• 2012

High-energy mechanical action applied to solid leads to destruction and diminution to the nanosize level. But on the other hand, it can induce structural changes at the nanoscale and at the atomic level which can result in novel materials properties. In this contribution, case studies will be presented concerned with the tailoring of magnetic properties of mechanically treated nanomaterials. Emphasis is placed on materials that have been synthesized by mechanochemical means and on an improved understanding of their nanomagnetism in general. The associated local structural changes of the iron containing magnetic materials discussed in the examples have been studied most suitably by $^{57}Fe$ Mossbauer nuclear probe spectroscopy whose results are supplemented by measurements of the magnetic properties of the mechanosynthesized nanomaterials.

• 한국초전도ㆍ저온공학회논문지
• /
• 제17권4호
• /
• pp.21-24
• /
• 2015

We report optical spectra of underdoped $BaFe_{1.89}Co_{0.11}As_2$ that hosts both of magnetic and superconducting orders. The temperature dependent evolution of optical conductivity shows finger prints of the magnetic order and the s-wave nature of the superconducting gaps. Careful inspection on the superconducting state reveals that the two orders compete but coexist in the ground state.

• 셀메드
• /
• 제3권1호
• /
• pp.7.1-7.7
• /
• 2013

Brassica juncea is a polyphenols enriched edible plant, with diverse medicinal uses of different parts of which have been mentioned in the Ayurveda. The effects of 10 daily oral doses (100, 200, and 400 mg/kg/day) of a methanolic Brassica juncea leaf extract in rat models of anxiety using nondiabetic and alloxan-diabetic rats were quantified. In all the three behavioural tests used, i.e. elevated plus maze, open field, and social interaction tests, anxiolytic-like activity of the extract was observed in the diabetic animals only. Quantitatively, the efficacy of the highest tested dose of the extract in these tests was always less than those observed after its lower ones. These observations provide further experimental evidences for the conviction that Brassica vegetables could as well be useful for combating diabetes associated mental health problems.