• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.128-128
• /
• 2019

증발산량은 수표면이나 토양면에서 수증기의 형태로, 대기중으로 방출되는 증발량과 식물의 엽면을 통해 지중의 물이 대기 중으로 방출되는 증산량의 합으로, 기상학과 수문학에 사용되는 중요한 농업기상 매개 변수이다. 증발산량을 관측하는 방법으로는 라이시미터 (Lysimeter)와 같은 관측장비를 통해 실제 증발산량을 측정하는 방법과, FAO-56 Penman-Monteith (PM)과 같은 증발산량 추정 알고리즘을 이용하여 산출하는 방법이 있다. 국내의 경우 기상관측소에서 수집한 데이터를 이용하여 증발산량 추정공식을 통해 증발산량을 산정하는 연구가 이루어졌으며, 위성영상에 기반하여 증발산을 추정하려는 연구가 진행되고 있다. 본 연구에서는 미국 항공우주국 (National Aeronautics and Space Administration, NASA)에서 추진하는 위성을 이용한 지구 전역의 장기관측 계획 EOS (Earth Observing System)에 의해 발사된 지구 관측 위성인 MODIS Terra 위성에서 제공되는 MOD16A2 위성영상을 사용하였다. MOD16A2 위성영상은 2001년부터 현재까지 500m의 픽셀 단위로 제공되는 8일 간격의 전지구 규모의 위성영상으로, 본 연구에서는 우리나라 관측소에서 관측된 기상인자를 PM 공식에 입력하여 산정된 증발산량 값과 MOD16A2 위성영상 데이터를 비교하여 우리나라 MOD16A2 위성영상 적용성 및 밭작물 가뭄분석에 적용하였다.

• Transactions on Semiconductor Engineering
• /
• v.1 no.1
• /
• pp.23-30
• /
• 2023

With the recent development of Generative AI technology by IT giants, the size of the transformer model is increasing exponentially over trillion won. In order to continuously enable these AI services, it is essential to reduce the weight of the model. In this paper, we find a hardware-friendly structured pruning pattern and propose a lightweight method of the transformer model. Since compression proceeds by utilizing the characteristics of the model algorithm, the size of the model can be reduced and performance can be maintained as much as possible. Experiments show that the structured pruning proposed when pruning GPT-2 and BERT language models shows almost similar performance to fine-grained pruning even in highly sparse regions. This approach reduces model parameters by 80% and allows hardware acceleration in structured form with 0.003% accuracy loss compared to fine-tuned pruning.

• Geotechnical Engineering
• /
• v.14 no.1
• /
• pp.93-108
• /
• 1998

In this study, the analytic solutions and numerical methods were used to estimate the shape and size effects on the stability of underground openings. The stability of underground openings was evaluated by scrutinizing the effects of the rock mass quality, the state of in-situ stresses and the lateral earth pressure coefficient on the displacement, the stress concentration and the plastic region developed in the vicinity of the openings. The analytic solutions have shown that the stress concentration factor is inversely proportional to the radius of curvature of openings. Through parametric study on the various shapes and sizes of underground openings the characteristics of the controlling factors concerned with the stability were analyzed. Then, the study was extended to the horseshoe-shaped openings commonly used for under ground storage. Through the extended study the effects of the stress ratio and the height-towidth ratio of openings on the maximum displacement and plastic region developed around the openings were estimated. The results have shorn that the height-to-width ratio of domestic storage caverns can be increased economically without stability problem, as far as the lateral earth pressure coefficient is appropriate.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.310-310
• /
• 2020

자연하천에서 식생의 성장은 유속 및 수위 변화에 중요한 영향을 미치며, 따라서 식생으로 인한 흐름저항은 흐름과 유사이송 모델링을 위한 중요 매개 변수가 된다. 즉, 수치 모델을 활용함에 있어 식생의 흐름 저항을 정확하게 추정하는 것이 매우 중요하며, 보다 정확한 추정을 위해 실험 혹은 현장 데이터를 활용한 보정과정이 필수적으로 요구된다. 본 연구에서는 식생 패치를 포함한 인공 수로의 흐름 모의를 위해 실규모 실험 수로에서 측정된 유속 및 수위 데이터를 활용하여 모델보정을 수행하고자 한다. 이를 위해 공간 분포 별로 각기 다른 흐름 저항식의 적용이 가능하며, 식생 저항 공식을 포함하고 있는 Delft3D 모델을 활용하였다. 또한 실규모 수로에서의 유속 및 수위 데이터 수집을 위해 한국건설기술연구원 하천연구센터에서 실험을 수행하였다. 실험 구간의 길이는 약 120 m이고 하폭은 11 m이며, 국내 하천에서 보이는 식생패치의 유사한 형태를 재현하기 위해 하천 내 가장 많이 활착되어있는 버드나무와 유사한 형태의 인공식생을 제작하였다. 인공 식생은 지그재그로 배치되었으며, 식생의 전체 높이는 1.1 m이고, 각 패치 당 23그루의 인공 식생이 총 8개 패치에 식재되었다. 모의 조건은 상류단 유입 유량 2.805 ㎥/s, 하류단수위 98.764 m의 정류 조건을 적용하였다. 또한 식생 패치 구간에서의 흐름 저항 추정을 위해 Delft3D 모델 내에서 선택가능 한 Baptist의 비침수(Non-submerged)식을 적용하였으며, 항력계수 결정을 위해 1과 1.5를 적용하여 측정 수위와 비교하였다. Delft3D의 흐름 모의 결과, 항력계수 1.5를 적용했을 때, 측정 수위와 거의 일치하는 것으로 나타났으며, 항력계수 1.0을 적용했을 경우, 측정 수위에 비해 다소 낮게 모의되는 것으로 나타났다. 또한 항력계수 1.5인 경우 식생 패치 구간에서 평균 0.65 m/s의 유속이 발생하였다.

• Journal of Korean Society of Steel Construction
• /
• v.14 no.5 s.60
• /
• pp.603-612
• /
• 2002

Numerical analyses have been carried out in order to check the applicability of high-strength steel to a medium-sized steel bridge. Using the yield strength of steel, Average Daily Truck Traffic (ADTT), and fatigue grade of structural detail as analytical parameters, the minimum weight sections that satisfy the limit states of the AASHTO LRFD design specification were determined through an optimization scheme. Likewise, the effects of the number of girders and span length on the applicability of high-strength steel were evaluated. Results show that high-strength steel may be employed in the steel bridge, since steel weight decreases with increasing yield strength regardless of the fatigue effect. Nonetheless, appropriate countermeasures against fatigue should be determined since it is a major factor in the effective use of high-strength steel in steel bridges.

• Tunnel and Underground Space
• /
• v.24 no.2
• /
• pp.155-165
• /
• 2014

Multiple thermal energy storage (TES) caverns can be used for storing thermal energy on a large scale and for a high-aspect-ratio heat storage design to provide good thermal performance. It may also be necessary to consider the use of multiple caverns with a reduced length when a single, long tunnel-shaped cavern is not suitable for connection to aboveground heat production and injection equipments. When using multiple TES caverns, the separation distance between the caverns is one of the significant factors that should be considered in the design of storage space, and the optimal separation distance should be determined based on a quantitative stability criterion. In this paper, we described a numerical approach for determining the optimal separation distance between multiple caverns for large-scale TES utilization. For reliable stability evaluation of multiple caverns, we employed a probabilistic method which can quantitatively take into account the uncertainty of input parameters by probability distributions, unlike conventional deterministic approaches. The present approach was applied to the design of a conceptual TES model to store hot water for district heating. The probabilistic stability results of this application demonstrated that the approach in our work can be effectively used as a decision-making tool to determine the optimal separation distance between multiple caverns. In addition, the probabilistic results were compared to those obtained through a deterministic analysis, and the comparison results suggested that care should taken in selecting the acceptable level of stability when using deterministic approaches.

• Journal of the Korea Concrete Institute
• /
• v.15 no.4
• /
• pp.606-614
• /
• 2003

The major cause of deterioration for the concrete bridge decks exposed to de-icing chemicals would be chloride-induced reinforcement corrosion. Thus, in this paper, in order to predict time to corrosion for concrete bridge decks in the urban area, chloride concentration was measured with depth from the surface. A frequency analysis on surface chloride concentration and chloride diffusion coefficient of concrete bridge deck equals 0.192, 29.828 in the scale parameter and 7.899, 1.983 in the shape parameter of gamma distribution. The average value of surface chloride concentration equals 1.5 kg/㎥ and condenses from 1 to 2 kg/㎥ in the level of probability 70%. From the probabilistic results, it is confirmed that 26mm of minimum cover depth in order to target 20 years over is calculated. The countermeasure strategy to extend the service life of concrete bridge deck exposed to de-icing chemicals would be an effective method to increase cover depth and to place high performance concrete, which could lead to reduce the chloride diffusion coefficient and distribution range.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.16 no.3
• /
• pp.193-210
• /
• 2013

This study aims to assess the slope stability variation of Jeollabuk-do drainage areas by RCM model outputs based on A1B climate change scenario and infinite slope stability model based on the specific catchment area concept. For this objective, we downscaled RCM data in time and space: from watershed scale to rain gauge scale in space and from monthly data to daily data in time and also developed the GIS-based infinite slope stability model based on the concept of specific catchment area to calculate spatially-distributed wetness index. For model parameterization, topographic, geologic, forestry digital map were used and model parameters were set up in format of grid cells($90m{\times}90m$). Finally, we applied the future daily rainfall data to the infinite slope stability model and then assess slope stability variation under the climate change scenario. This research consists of two papers: the first paper focuses on the methodologies of climate change scenario preparation and infinite slope stability model development.

• International Commerce and Information Review
• /
• v.16 no.5
• /
• pp.343-359
• /
• 2014

This study investigates how cultural differences between countries affect bilateral trade volumes, using Hofstede's cultural index that reflects nations' cultural characteristics. Empirical analyses of the impacts of Hofstede's five cultural characteristics on bilateral trade volumes are conducted either in each separate equation or simultaneously. Bilateral trade data of OECD countries plus China as of year 2010 is used for regression analysis on gravity model. Regression results from individual equation for each cultural index variable show tthe smaller the index gaps of power distances and uncertainty avoidance among countries, the larger bilateral trade volumes. On the contrary, the larger the index gaps of long-term orientation among countries, the larger bilateral trade volumes. If five Hofstede cultural indexes are regressed in a single equation, however, only variables of power distance and long-term orientation are significant. The analysis largely confirms that bilateral trade among countries with similar culture have much potrential to grow. It implies that policy actions for cultural proximity are very important for furthering bilateral trade.

• Journal of Internet Computing and Services
• /
• v.19 no.1
• /
• pp.77-86
• /
• 2018

For the past few years, KISTI has been servicing an online simulation execution platform, called EDISON, allowing users to conduct simulations on various scientific applications supplied by diverse computational science and engineering disciplines. Typically, these simulations accompany large-scale computation and accordingly produce a huge volume of output data. One critical issue arising when conducting those simulations on an online platform stems from the fact that a number of users simultaneously submit to the platform their simulation requests (or jobs) with the same (or almost unchanging) input parameters or files, resulting in charging a significant burden on the platform. In other words, the same computing jobs lead to duplicate consumption computing and storage resources at an undesirably fast pace. To overcome excessive resource usage by such identical simulation requests, in this paper we introduce a novel framework, called IceSheet, to efficiently manage simulation data based on execution metadata, that is, provenance. The IceSheet framework captures and stores each provenance associated with a conducted simulation. The collected provenance records are utilized for not only inspecting duplicate simulation requests but also performing search on existing simulation results via an open-source search engine, ElasticSearch. In particular, this paper elaborates on the core components in the IceSheet framework to support the search and reuse on the stored simulation results. We implemented as prototype the proposed framework using the engine in conjunction with the online simulation execution platform. Our evaluation of the framework was performed on the real simulation execution-provenance records collected on the platform. Once the prototyped IceSheet framework fully functions with the platform, users can quickly search for past parameter values entered into desired simulation software and receive existing results on the same input parameter values on the software if any. Therefore, we expect that the proposed framework contributes to eliminating duplicate resource consumption and significantly reducing execution time on the same requests as previously-executed simulations.