• Title/Summary/Keyword: 결정성 분석

Search Result 11,207, Processing Time 0.042 seconds

Estimation of River Flow Data Using Machine Learning (머신러닝 기법을 이용한 유량 자료 생산 방법)

  • Kang, Noel;Lee, Ji Hun;Lee, Jung Hoon;Lee, Chungdae
    • 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.

  • PDF

Effects of 4 major river project on the flood management in Yeongsan-river basin (영산강유역 홍수관리에서 4대강 사업의 효과)

  • Song, Jin Keun;Jung, Jae Sung
    • Proceedings of the Korea Water Resources Association Conference
    • /
    • 2021.06a
    • /
    • pp.301-301
    • /
    • 2021
  • 영산강 유역은 상류 급경사 지역에 있는 유역 내 댐들의 홍수조절용량과 평야지역에 있는 하굿둑의 홍수배제능력이 부족하여 나주시 등 하류부에 홍수피해가 빈번하였다. 영산강에서 4대강 사업의 주요내용은 하도내 퇴적토준설과 홍수조절지 건설, 농업용저수지 증고를 통한 홍수방어능력 증대, 다기능 보설치를 통한 용수확보, 하구지역 홍수배제능력 증대 등이었다. 동 사업에서 본류 하천의 홍수위 저감과 내수배제 개선을 위해 담양댐 하류부터 영암천 합류점까지 하도를 준설하고, 담양과 화순홍수조절지, 나주 강변저류지, 승촌보와 죽산보를 건설하고, 하굿둑 배수능력 증대와 농업용 댐 및 저수지들을 증고를 수행하였다. 4대강 사업이 준공된 2012년 이후로 2014년부터 5년간 가뭄이 지속되었고 큰 홍수가 없다가 2020년 8월에 장기간의 집중호우로 영산강 중상류인 광주시와 영산강 하류지역에서 큰 수해가 발생하였다. 따라서 시설물 운영 실적에 근거한 홍수저감 효과의 기술적 검토를 수행하였다. 사업전·후 수문관측자료와 하천시설 운영 실적에 근거한 홍수저감 효과를 분석하기 위해 사업전·후 유사 규모의 강우 발생 시 수위표 지점별로 계측된 첨두 수위 및 유량자료를 비교하여 홍수저감 효과를 분석하였다. 사업전·후 유사 규모 강우를 선정하기 위해 발생된 강우 사상 중 호우특보 발령 기준이상의 강우 사상에 대하여 총 강우량 및 강우의 지속시간, 시간 분포를 비교하여 유사 규모의 호우를 선정하였다. 사업전·후 유사 규모의 호우 사상 발생 시 계측된 홍수위와 홍수량 비교 결과 영산강 중·상류부와 중·하류부 수위표 지점(극락교,승용교,나주대교)에서 사업 시행 후 사업 전보다 첨두 수위가 1.36~2.81m 감소한 것으로 검토되었다. 이는 여러 가지 사업들의 복합적인 결과로 영산강유역의 홍수관리여건이 개선된 것으로 판단된다. 한편 2020년 8월7일~8일 발생한 호우에 의해 영산강 본류의 중·상류부와 중·하류부의 주요 수위표 지점에서는 200년빈도 계획홍수위를 초과한 홍수가 발생하였다. 상시개방과 철거로 처리방안이 결정된 승촌보와 죽산보의 여건을 반영하여 2개보 유무에 따른 홍수위 검토를 실시하였다. 홍수위 비교 결과보가 없을 경우 영산강 중·상류부(극락교,승용교)와 중·하류부(나주대교,영산교) 수위표 지점에서 홍수위가 0.01~0.07m 감소되는 것으로 검토되어 홍수시 보의 영향은 비교적 작은 것으로 나타났다. 홍수시 상류댐과 저수지, 홍수조절지, 하굿둑, 하천의 연계운영에 대해서는 추가연구가 필요하다.

  • PDF

Application of Intensity-Duration-Frequency Curve to Korea Derived by Cumulative Distribution Function (누가분포함수를 활용한 강우강도식의 국내 적용성 평가)

  • Kim, Kewtae;Kim, Taesoon;Kim, Sooyoung;Heo, Jun-Haeng
    • KSCE Journal of Civil and Environmental Engineering Research
    • /
    • v.28 no.4B
    • /
    • pp.363-374
    • /
    • 2008
  • Intensity-Duration-Frequency (IDF) curve that is essential to calculate rainfall quantiles for designing hydraulic structures in Korea is generally formulated by regression analysis. In this study, IDF curve derived by the cumulative distribution function ("IDF by CDF") of the proper probability distribution function (PDF) of each site is suggested, and the corresponding parameters of IDF curve are computed using genetic algorithm (GA). For this purpose, IDF by CDF and the conventional IDF derived by regression analysis ("IDF by REG") were computed for 22 Korea Meteorological Administration (KMA) rainfall recording sites. Comparisons of RMSE (root mean squared error) and RRMSE (Relative RMSE) of rainfall intensities computed from IDF by CDF and IDF by REG show that IDF by CDF is more accurate than IDF by REG. In order to accommodate the effect of the recent intensive rainfall of Korea, the rainfall intensities computed by the two IDF curves are compared with that by at-site frequency analysis using the rainfall data recorded by 2006, and the result from IDF by CDF show the better performance than that from IDF by REG. As a result, it can be said that the suggested IDF by CDF curve would be the more efficient IDF curve than that computed by regression analysis and could be applied for Korean rainfall data.

Comparison of WBCs and RBCs Concordance between the Cobas u 701 Analyzer and Three Manual Microscopy Methods in Urine Sediment (소변 침전물에서 Cobas u 701 Analyzer와 세 가지 수동 현미경검사법의 백혈구와 적혈구의 일치도 비교)

  • Hyeok-Jae LEE;Min-Hyeok LEE
    • Korean Journal of Clinical Laboratory Science
    • /
    • v.55 no.2
    • /
    • pp.82-92
    • /
    • 2023
  • Urine sediments are performed by a microscopic examination of centrifuged urine by medical technologists. This study examined different urine sediment preparation procedures. The 107 fresh urine specimens that tested positive from white blood cells (WBCs) and red blood cells (RBCs) in the urine dipstick test and the cobas u 701 analyzer, respectively, were selected for manual microscopy. This study evaluated an automated urine sediment analyzer and three manual microscopy methods for WBCs and RBCs. The methods were performed according to the test guidelines. The coefficients of determination between the cobas u 701 analyzer and the Korean Association of Quality Assurance for Clinical Laboratory (KAQACL) for WBCs and RBCs were r2=0.977 and r2=0.970, respectively. The concordance rates between the cobas u 701 analyzer and KAQACL for WBCs and RBCs were 74.8% and 77.6%, respectively. A good correlation and concordance with the automatic analyzer were shown when the specimens were prepared and examined using the KAQACL method. Consequently, the differences in the urine sediment preparation procedures affected the sediment concentrations, influencing the cell number per high power field (HPF).

A Modified grid-based KIneMatic wave STOrm Runoff Model (ModKIMSTORM) (II) - Application and Analysis - (격자기반 운동파 강우유출모형 KIMSTORM의 개선(II) - 적용 및 분석 -)

  • Jung, In Kyun;Shin, Hyung Jin;Park, Jin Hyeog;Kim, Seong Joon
    • KSCE Journal of Civil and Environmental Engineering Research
    • /
    • v.28 no.6B
    • /
    • pp.709-721
    • /
    • 2008
  • This paper is to test the applicability of ModKIMSTORM (Modified KIneMatic Wave STOrm Runoff Model) by applying it to Namgangdam watershed of $2,293km^2$. Model inputs (DEM, land use, soil related information) were prepared in 500 m spatial resolution. Using five typhoon events (Saomi in 2000, Rusa in 2002, Maemi in 2003, Megi in 2004 and Ewiniar in 2006) and two storm events (May of 2003 and July of 2004), the model was calibrated and verified by comparing the simulated streamflow with the observed one at the outlet of the watershed. The Pearson's coefficient of determination $R^2$, Nash and Sutcliffe model efficiency E, the deviation of runoff volumes $D_v$, relative error of the peak runoff rate $EQ_p$, and absolute error of the time to peak runoff $ET_p$ showed the average value of 0.984, 0.981, 3.63%, 0.003, and 0.48 hr for 4 storms calibration and 0.937, 0.895, 8.08%, 0.138, and 0.73 hr for 3 storms verification respectively. Among the model parameters, the stream Manning's roughness coefficient was the most sensitive for peak runoff and the initial soil moisture content was highly sensitive for runoff volume fitting. We could look into the behavior of hyrologic components from the spatial results during the storm periods and get some clue for the watershed management by storms.

Stability Analysis of Unsaturated Weathered Soil Slopes Considering Rainfall Duration (지속강우특성에 따른 불포화 풍화토사면의 안정성분석)

  • Jeong, Sang-Seom;Choi, Jae-Young;Lee, Jae-Hwan
    • KSCE Journal of Civil and Environmental Engineering Research
    • /
    • v.29 no.1C
    • /
    • pp.1-9
    • /
    • 2009
  • In this study, the influence of wetting band depth by continuous rainfall and the magnitude of wetting front suction on the stability of slopes in weathered soils were investigated by using finite element programs SEEP/W and SLOPE/W. Three different intensities of rainfall (10mm/hr, 30mm/hr, 50mm/hr) were chosen, and the total duration of rainfall was 96 hours. Three infinite slopes with the inclination of 1:1.5 and 1:1.8, 1:2.0 were considered and the typical properties and the shear strength parameters of the weathered soil were applied. It is shown that rainfall duration plays an important role in slope stability. Based on the analytical results, it is found that as the rainfall duration increases, the wetting band depth also increases. Also, the increasing rate of the wetting band depth was decreased as the soil density was increased. These results come from the decrease of the coefficient of permeability and the increase of the soil suction. Finally, it is also shown that the safety factors of slopes by unsaturated analysis are mostly larger than those by saturated analysis. Therefore, commonly used saturated analysis may substantially underestimate the degree of safety factor in realistic situations.

Residual Pesticide Analysis Method of Edible Oil via Heat Distillation Methods (가열증류법에 의한 식용유지의 잔류농약 분석법 개발)

  • Mi-Hui Son;Jae-Kwan Kim;Young-Seon Cho;Na-Eun Han;Byeong-Tae Kim;Myoung-Ki Park;Yong-Bae Park
    • Journal of Food Hygiene and Safety
    • /
    • v.38 no.3
    • /
    • pp.89-98
    • /
    • 2023
  • Currently, no guidelines exist regarding the maximum residues of pesticides in edible oil which is a processed food commonly consumed in Korea. This lack of guidelines hinders the evaluation of the safety of edible oil in terms of pesticide contamination. In this study, an analysis method based on heat distillation and GC-MS/MS was established by optimizing the extraction and purification procedure for 68 pesticides. Important variables in the thermal distillation procedure included heating temperature and time, and we found the nitrogen flow rate as a mobile phase and the type of dissolving solvent were not considerably affected. The determination coefficient (R2) of the residual pesticide was 0.99 or higher, and the quantitative limit (LOQ) was 0.01-0.02 mg/L. The average recovery rate (n=5) was 66.1-120.0% and the relative standard deviation was lower than ±10% when 68 pesticides were spiked at concentrations of 0.01-0.02, 0.1, and 0.5 mg/L. In addition, the within-laboratory precision was less than ±11%, meeting the Korea Food and Drug Safety Evaluation Institute's Guidelines on Standard Procedures for Preparing Food Testing Methods (2016). Therefore, the test method developed in this study can be used as a test method for managing the safety of the residual pesticide concentration in edible oil.

Factors Associated with Body Mass Index(BMI) Among Older Adults: A Comparison Study of the U.S., Japan, and Korea (노인의 체질량지수에 관련된 요인 연구: 미국, 일본, 한국 비교를 중심으로)

  • Yeom, Jihye;Kim, Jung Ki;Crimmins, Eileen M.
    • 한국노년학
    • /
    • v.29 no.4
    • /
    • pp.1479-1500
    • /
    • 2009
  • This study examined BMI distributions among older adults in three different countries: the U.S., Japan, and Korea. The paper also explored differences in the factors predicting BMI in the three countries using three data sets: the U.S. Longitudinal Study of Aging (LSOA II, 8,589 persons), the Nihon University Japanese Longitudinal Study of Aging (NUJLSOA, 2,888 persons), and the Korean Longitudinal Study of Ageing (KLoSA, 2,397 persons). Descriptive analysis and multiple regression were performed. Japanese older adults were somewhat lighter than Koreans with fewer people at the upper end of the BMI distribution. Distributions of BMI among both Koreans and Japanese are shifted leftward relative to Americans. There is less dispersion in the distribution of BMI for Koreans and Japanese than among Americans. The association between socioeconomic variables and BMI is stronger in the U.S. and Japan than in Korea. Demographic variables are strong predictors of BMI in Korea. In Japan, all health behaviors have significant effects on BMI. It is concluded that the relationships between behavioral, demographical, and socioeconomic factors and BMI are not the same across countries. Results have policy implications for the involvement of health practitioners in helping older adults to control weight.

The Effect of Perceived Customer Value on Customer Satisfaction with Airline Services Using the BERTopic Model (BERTopic 모델을 이용한 항공사 서비스에서 지각된 고객가치가 고객 만족도에 미치는 영향 분석)

  • Euiju Jeong;Byunghyun Lee;Qinglong Li;Jaekyeong Kim
    • Knowledge Management Research
    • /
    • v.24 no.3
    • /
    • pp.95-125
    • /
    • 2023
  • As the aviation industry has rapidly been grown, there are more factors for customers to consider when choosing an airline. In response, airlines are trying to increase customer value by providing high-quality services and differentiated experiential value. While early customer value research centered on utilitarian value, which is the trade-off between cost and benefit in terms of utility for products and services, the importance of experiential value has recently been emphasized. However, experiential value needs to be studied in a specific context that fully represents customer preferences because what constitutes customer value changes depending on the product or service context. In addition, customer value has an important influence on customers' decision-making, so it is necessary for airlines to accurately understand what constitutes customer value. In this study, we collected customer reviews and ratings from Skytrax, a website specializing in airlines, and utilized the BERTopic technique to derive factors of customer value. The results revealed nine factors that constitute customer value in airlines, and six of them are related to customer satisfaction. This study proposes a new methodology that enables a granular understanding of customer value and provides airlines with specific directions for improving service quality.

Characteristics and Standards of Domestic Tidal Flat Mud Marine Healing Resources (국내 갯벌머드 해양치유자원의 특성 및 기준에 관한 연구)

  • Seonyoung Park;Jeongwon Kang;Yonggi Jeong;Yeonje Cho
    • Journal of Wetlands Research
    • /
    • v.25 no.4
    • /
    • pp.386-393
    • /
    • 2023
  • The domestic marine healing industry is undergoing significant revitalization efforts, with a focus on understanding the efficacy and effectiveness of marine healing resources. This study establishes utilization and management standards through a detailed analysis of the active components within well-recognized marine healing mud materials. Samples of mud materials were collected from domestic tidal flats. These samples exhibited an average composition of 7.87% sand, 74.95% silt, and 17.17% clay, with a combined mud content (silt+clay)(silt+clay) consistently exceeding 90%. Notably, SiO2 emerged as the most prevalent effective ingredient at 68.4%, followed by Al2O3 (13.3%)>Fe2O3 (4.0%)>K2O (2.9%)>Na2O (2.3%)>MgO (1.6%)>CaO (1.0%)>TiO2 (0.7%), in terms of average content. Subsequently, through an analysis of effective ingredients, Si, Al, Fe, K, Na, Mg, and Ca were identified as elements demonstrating significant functionality. Among these, key indicator ingredients were selected for quality control, all of which were found to possess efficacious properties. Notably, K, Mg, and Ca exhibited particularly high concentrations. Based on these findings and referencing existing literature, it is recommended that domestic tidal flat mud resources earmarked for utilization as marine healing resources should possess a raw material mud content of no less than 70.0%. Moreover, the cumulative index components K2O+MgO+CaO should meet or exceed a threshold of 5.0% for optimal effectiveness.