• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.261-261
• /
• 2020

물관리의 기본이 되는 연속적인 유량 자료 확보를 위해서는 정확도 높은 수위-유량 관계 곡선식 개발이 필수적이다. 수위-유량 관계곡선식은 모든 수문시설 설계의 기초가 되며 홍수, 가뭄 등 물재해 대응을 위해서도 중요한 의미를 가지고 있다. 그러나 일반적으로 유량 측정은 많은 비용과 시간이 들고, 식생성장, 단면변화 등의 통제특성(control)이 변함에 따라 구간분리, 기간분리와 같은 비선형적인 양상이 나타나 자료 해석에 어려움이 존재한다. 특히, 국내 하천의 경우 자연적 및 인위적인 환경 변화가 다양하여 지점 및 기간에 따라 세밀한 분석이 요구된다. 머신러닝(Machine Learning)이란 데이터를 통해 컴퓨터가 스스로 학습하여 모델을 구축하고 성능을 향상시키는 일련의 과정을 뜻한다. 기존의 수위-유량 관계곡선식은 개발자의 판단에 의해 데이터의 종류와 기간 등을 설정하여 회귀식의 파라미터를 산출한다면, 머신러닝은 유효한 전체 데이터를 이용해 스스로 학습하여 자료 간 상관성을 찾아내 모델을 구축하고 성능을 지속적으로 향상 시킬 수 있다. 머신러닝은 충분한 수문자료가 확보되었다는 전제 하에 복잡하고 가변적인 수자원 환경을 반영하여 유량 추정의 정확도를 지속적으로 향상시킬 수 있다는 이점을 가지고 있다. 본 연구는 머신러닝의 대표적인 알고리즘들을 활용하여 유량을 추정하는 모델을 구축하고 성능을 비교·분석하였다. 대상지역은 안정적인 수량을 확보하고 있는 한강수계의 거운교 지점이며, 사용자료는 2010~2018년의 시간, 수위, 유량, 수면폭 등 이다. 프로그램은 파이썬을 기반으로 한 머신러닝 라이브러리인 사이킷런(sklearn)을 사용하였고 알고리즘은 랜덤포레스트 회귀, 의사결정트리, KNN(K-Nearest Neighbor), rgboost을 적용하였다. 학습(train) 데이터는 입력자료 종류별로 조합하여 6개의 세트로 구분하여 모델을 구축하였고, 이를 적용해 검증(test) 데이터를 RMSE(Roog Mean Square Error)로 평가하였다. 그 결과 모델 및 입력 자료의 조합에 따라 3.67~171.46로 다소 넓은 범위의 값이 도출되었다. 그 중 가장 우수한 유형은 수위, 연도, 수면폭 3개의 입력자료를 조합하여 랜덤포레스트 회귀 모델에 적용한 경우이다. 비교를 위해 동일한 검증 데이터를 한국수문조사연보(2018년) 내거운교 지점의 수위별 수위-유량 곡선식을 이용해 유량을 추정한 결과 RMSE가 3.76이 산출되어, 머신러닝이 세분화된 수위-유량 곡선식과 비슷한 수준까지 성능을 내는 것으로 확인되었다. 본 연구는 양질의 유량자료 생산을 위해 기 구축된 수문자료를 기반으로 머신러닝 기법의 적용 가능성을 검토한 기초 연구로써, 국내 효율적인 수문자료 측정 및 수위-유량 곡선 산출에 도움이 될 수 있을 것으로 판단된다. 향후 수자원 환경 및 통제특성에 영향을 미치는 다양한 영향변수를 파악하기 위해 기상자료, 취수량 등의 입력 자료를 적용할 필요가 있으며, 머신러닝 내 비지도학습인 딥러닝과 같은 보다 정교한 모델에 대한 추가적인 연구도 수행되어야 할 것이다.

• Journal of Wetlands Research
• /
• v.21 no.1
• /
• pp.34-42
• /
• 2019

Baseflow which is one of the unmeasurable components of streamflow and slowly flows through underground is important for water resource management. Despite various separation methods from researches preceded, it is difficult to find a significant separation method for baseflow separation. This study applied the MRC method and developed the improved approach to separate baseflow from total streamflow hydrograph. Previous researchers utilized the whole streamflow data of study period at once to derive synthetic MRCs causing unreliable results. This study has been proceeded with total nine areas with gauging stations. Each three areas are selected from 3 domestic major watersheds. Tool for drawing MRC had been used to draw MRCs of each area. First, synthetic MRC for whole period and two other MRCs were drawn following two different criteria. Two criteria were set by different conditions, one is flow condition and the other is seasonality. The whole streamflow was classified according to seasonality and flow conditions, and MRCs had been drawn with a specialized program. The MRCs for flow conditions had low R2 and similar trend to recession segments. On the other hand, the seasonal MRCs were eligible for the baseflow separation that properly reflects the seasonal variability of baseflow. Comparing two methods of assuming MRC for baseflow separation, seasonal MRC was more effective for relieving overestimating tendency of synthetic MRC. Flow condition MRCs had a large distribution of the flow and this means accurate MRC could not be found. Baseflow separation using seasonal MRC is showing more reliability than the other one, however if certain technique added up to the flow condition MRC method to stabilize distribution of the streamflow, the flow conditions method could secure reliability as much as seasonal MRC method.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.8 no.2
• /
• pp.102-114
• /
• 2004

The present work has been developed the interpretation processor including the behavior of material failure and the separation phenomena under transient dynamic loading (the operation of explosive bolt) using AUTODYN V4.3, SoildWork 2003 and TrueGrid V2.1 programs. It has been demonstrated that the interpretation in ridge-cut explosive bolt under dynamic loading condition should be necessary to the appropriate failure model and the basic stress of bolt failure is the principal stress. The use of this interpretation processor developing the present work could be extensively helped to design the shape and the amount of explosives in the explosive bolt having a complex geometry. It is also proved that the interpretation processor approach is an accurate and effective analysis technique to evaluate the separation mechanism in explosive bolts.

• Journal of Korea Water Resources Association
• /
• v.42 no.9
• /
• pp.725-736
• /
• 2009

A flood event can be defined by three characteristics; peak discharge, total flood volume, and flood duration, which are correlated each other. However, a conventional flood frequency analysis for the hydrological plan, design, and operation has focused on evaluating only the amount of peak discharge. The interpretation of this univariate flood frequency analysis has a limitation in describing the complex probability behavior of flood events. This study proposed a bivariate flood frequency analysis using a Gumbel mixed model for the flood evaluation. A time series of annual flood events was extracted from observations of inflow to the Soyang River Dam and the Daechung Dam, respectively. The joint probability distribution and return period were derived from the relationship between the amount of peak discharge and the total volume of flood runoff. The applicability of the Gumbel mixed model was tested by comparing the return periods acquired from the proposed bivariate analysis and the conventional univariate analysis.

• Journal of Life Science
• /
• v.14 no.1
• /
• pp.98-103
• /
• 2004

Various bacterial strains were isolated from water in a paddy field, and their seaweed degrading activities were investigated. As the result, 16 strains were obtained from the microflora sample. They were incubated in a liquid medium of sea tangle powder for 3 weeks. Ratios of reduced sugar to total sugar of the liquid medium were measured once a week during the incubation period. Ratio of reduced sugar to total sugar of 30A412 strain was highest. Accordingly, 30A412 strain was incubated in three different liquid media of sodium alginate, sea tangle powder, and sea mustard powder for 3 or 4 weeks. The ratios of reduced sugar to total sugar and cell growth were measured once a week during the incubation period. Ratios of reduced sugar to total sugar and cell growth were increased with the incubation period. Cell growth and degrading ratio were highest in the liquid medium with sea mustard powder.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.43 no.8
• /
• pp.739-747
• /
• 2015

With rapid growth in the technologies and demand of Remotely Piloted Aircraft Systems (RPASs), integration of such systems into the existing airspace system is becoming an issue in many countries. RPAS have different flight performances, communication characteristics, separation assurance mechanisms, and human machine interfaces from manned aircraft. To establish rules and regulations for RPAS integration, it is important to understand the impacts of RPASs on the airspace system. A simulation system that integrates manned aircraft, air traffic control, and RPASs is developed in Inha University to investigate these impacts through Human-in-The-Loop (HiTL) simulations. Three conflict resolution scenarios between a manned aircraft and a Remotely Piloted Aircraft (RPA) were constructed and tested. Human factors such as the response times of pilots and controllers were measured and analyzed as well as the risk of each maneuver.

• Korean Journal of Food Science and Technology
• /
• v.27 no.3
• /
• pp.298-302
• /
• 1995

This study was performed to know the availability of low erucic acid rapeseed oil (LEAR oil) for vegetable oil in commercial mayonnaise preparation. Mixed oils (blended soybean oil with LEAR oil) were prepared and mayonnaises were prepared with these oils and then compared the quality characteristics of these samples. The oxidative stability of vegetable oil with Rancimat test showed that LEAR oil was higher stability than soybean oil, and became higher with increase of LEAR oil ratio. Quality characteristics of mayonnaises with these oils showed that mayonnaise with mixed oils was preferred to that with LEAR oil itself. Furthermore, the quality characteristics of mayonnaises with mixed oils of soybean oil $(20{\sim}60%)$ plus LEAR oil $(80{\sim}40%)$ were better than those of mayonnaise with soybean oil solely.

• Food Science and Preservation
• /
• v.14 no.5
• /
• pp.516-525
• /
• 2007

The effect on quality improvement of Chokochujang by the addition of green tea extract (GT) was evaluated on the basis of total cell numbers, pH, acidity, sugar content and sensory evaluation during its storage at $25^{\circ}C$ and $37^{\circ}C$ for 3 weeks. A dorminant strain was isolated from the Chokochujang and identified as Bacillus amyloliquefaciens ER282 based on the 16S rDNA sequence. The ethanol extract from GT was investigated for its antibacterial activity aid showed good activities enough to use for Chokochujang as a natural preservative. Growth of microorganisms was significantly inhibited by adding GT extract to Chokochujang but pH, acidity, sugar content and color was not much changed. Sensory evaluation and overall preference, however, was continuously declined as storage period increased, whereas the addtion of 3% GT extract could delay the quality loss of Chokochujang during storage at $25^{\circ}C$ and $37^{\circ}C$ for 3 weeks.

• Journal of Korea Port Economic Association
• /
• v.28 no.4
• /
• pp.193-207
• /
• 2012

The of purpose of this study is to predict export and import container volumes using a Decision Tree analysis. Factors which can influence the volume of container cargo are selected as independent variables; producer price index, consumer price index, index of export volume, index of import volume, index of industrial production, and exchange rate(won/dollar). The period of analysis is from january 2002 to December 2011 and monthly data are used. In this study, CRT(Classification and Regression Trees) algorithm is used. The main findings are summarized as followings. First, when index of export volume is larger than 152.35, monthly export volume is predicted with 858,19TEU. However, when index of export volume is between 115.90 and 152.35, monthly export volume is predicted with 716,582TEU. Second, when index of import volume is larger than 134.60, monthly import volume is predicted with 869,227TEU. However, when index of export volume is between 116.20 and 134.60, monthly import volume is predicted with 738,724TEU.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.37 no.6
• /
• pp.1-11
• /
• 2000

In this paper, we propose a novel control signaling protocol that efficiently controls connection setup and increases the utilization of network resources. The proposed protocol, Separated Control Signaling Protocol(SCSP), separates bearer control from call control lot WDM optical networks. The main function of call control is to check the availability of network resources such as wavelengths and receivers at destination node. Bearer control is to reserve and assign wavelengths. The signaling architecture of this protocol consists of call controller and hearer controller The call controller handles call setup and release, activates the beater controller, and manages the status of call and bearer. The bearer controller reserves wavelengths, sets up bearer, tears down bearer. and notifies the status of beater to call controller. The state transition diagrams of each controller are designed. Using control messages and related primitives, the information flows for call setup and bearer setup, hearer teardown and call release, and reaction for setup failures are described to evaluate the performance. The simulation results show that the separated control signaling protocol is superior to conventional one in terms of call blocking probability and resource utilization.