• Journal of Korea Water Resources Association
• /
• v.45 no.12
• /
• pp.1321-1330
• /
• 2012

In this study, the daily inflow at the basin of Chungju dam is predicted using wavelet-artificial neural network for nonlinear model. Time series generally consists of a linear combination of trend, periodicity and stochastic component. However, when framing time series model through these data, trend and periodicity component have to be removed. Wavelet transform which is denoising technique is applied to remove nonlinear dynamic noise such as trend and periodicity included in hydrometeorological data and simple noise that arises in the measurement process. The wavelet-artificial neural network (WANN) using data applied wavelet transform as input variable and the artificial neural network (ANN) using only raw data are compared. As a results, coefficient of determination and the slope through linear regression show that WANN is higher than ANN by 0.031 and 0.0115 respectively. And RMSE and RRMSE of WANN are smaller than those of ANN by 37.388 and 0.099 respectively. Therefore, WANN model applied in this study shows more accurate results than ANN and application of denoising technique through wavelet transforms is expected that more accurate predictions than the use of raw data with noise.

• Journal of Korea Water Resources Association
• /
• v.47 no.2
• /
• pp.135-143
• /
• 2014

In this study, Chungju inflow was simulated for climate change considering the uncertainties of GCMs and a stochastic model. TFN (Transfer Function Noise) model and 4 different GCMs (CNRM, CSIRO, CONS, UKMO) based on IPCC AR4 A2 scenario were used. In order to evaluate uncertainty of TFN model, 100 cases of noises are applied to the TFN model. Thus, 400 cases of inflow results are simulated. Future inflows according to the GCMs show different rates of changes for the future 3 periods relative to the past 30-years reference period. As the results, the summer inflow shows increasing trend and the spring inflow shows decreasing trend based on AR4 A2 scenario.

• Journal of Korea Water Resources Association
• /
• v.42 no.10
• /
• pp.877-889
• /
• 2009

This study is to assess the future potential impact of climate change on stream water quality for a 6,581.1 km$^2$ dam watershed using SWAT (Soil and Water Assessment Tool) model. The ECHAM5-OM climate data of IPCC (The Intergovernmental Panel on Climate Change) A2, A1B, and B1 emission scenarios were adopted and the future data (2007-2099) were corrected using 30 years (1977-2006, baseline period) weather data and downscaled by Change Factor (CF) method. After model calibration and validation using 6 years (1998-2003) observed daily streamflow and monthly water quality (SS, T-N, and T-P) data, the future (2020s, 2050s and 2080s) hydrological behavior and stream water quality were projected.

• Journal of Korea Water Resources Association
• /
• v.39 no.10 s.171
• /
• pp.845-854
• /
• 2006

Establishment of snowmelt factors is necessary to simulate stream flow using snowmelt models during snowmelt periods. The few observed data related snowmelt was the major cause of difficulty in extracting snowmelt factors such as snow cover area, snow depth and depletion curve. The objective of this study was to extract snowmelt factors using RS, GIS technique and meteorological data. Snow cover maps were derived from NOAA/AVHRR images for the winter seasons from 1997 to 2003. Distributed snow depth was mapped by overlapping between snow cover maps and interpolated snowfall maps from 69 meteorological observation station. Depletion curves of snowmelt area were described from the linear regression equations of each year between the average temperature and snow cover area in Soyanggang-dam and chungju-dam watershed.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.37 no.1
• /
• pp.231-238
• /
• 2017

Water supply safety index plays an important role on assessing the water supply capacity of hydrologic system. Due to the absence of consistent guidance, however, practical problems have been brought up on data period used for dam design and performance evaluation. Therefore, this study employed bivariate drought frequency analysis which is able to consider drought severity and duration simultaneously, in order to evaluate water supply capacity of multi-purpose dams. Drought characteristics were analyzed based on the probabilistic approach, and water supply capacity of five multi-purpose dams in Korea (Soyang River, Chungju, Andong, Daecheong, Seomjin River) were evaluated under the specific drought conditions. As a result, it would be possible to have stable water supply with their own inflow during summer and fall, whereas water shortage would occur even under the 1-year return period drought event during spring and winter due to low rainfall.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2007.05a
• /
• pp.1053-1057
• /
• 2007

본 연구에서는 준분포형 물리모형인 SWAT 모형을 통하여 유역 유사량 및 오염부하량 발생의 시공간적 특성을 파악하기 위해 충주댐 상류유역을 대상으로 모형을 구축 적용하였다. 대상유역에 대한 유출 유사 영양물질 관련 매개변수의 보정 및 모형 검증을 수행한 결과, 유출에 대해서는 모형효율지수 0.8 정도의 안정적인 결과를 얻을 수 있었으며, 유사와 인에 대해서는 대략적인 정성적 경향을 파악할 수 있었다. 구축된 모형을 통하여 대상유역에 대해 배수면적별 비유사량의 일정한 관계를 도출할 수 있었으며, 식생에 따른 단위면적별 발생 유사량 및 오염부하량을 검토함으로써 식생별 침식 및 부하량 발생 정도를 평가할 수 있었다. 또한, 하도구간별 오염원에 따른 유사 및 오염부하량을 검토하고, 월별 평균 유사량과 오염부하량을 검토함으로써 시공간적인 분포 특성을 파악할 수 있었다. 아울러 유역내 토양보전기법의 적용에 따른 유사 및 총질소, 총인의 저감효과도 평가할 수 있었다.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.59 no.3
• /
• pp.97-110
• /
• 2017

This study investigates the efficiencies of machine learning models, including artificial neural network (ANN), generalized regression neural network (GRNN), adaptive neuro-fuzzy inference system (ANFIS) and random forest (RF), for reservoir water level forecasting in the Chungju Dam, South Korea. The models' efficiencies are assessed based on model efficiency indices and graphical comparison. The forecasting results of the models are dependent on lead times and the combination of input variables. For lead time t = 1 day, ANFIS1 and ANN6 models yield superior forecasting results to RF6 and GRNN6 models. For lead time t = 5 days, ANN1 and RF6 models produce better forecasting results than ANFIS1 and GRNN3 models. For lead time t = 10 days, ANN3 and RF1 models perform better than ANFIS3 and GRNN3 models. It is found that ANN model yields the best performance for all lead times, in terms of model efficiency and graphical comparison. These results indicate that the optimal combination of input variables and forecasting models depending on lead times should be applied in reservoir water level forecasting, instead of the single combination of input variables and forecasting models for all lead times.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.51 no.2
• /
• pp.7-14
• /
• 2009

The purpose of this study is to identify how much the MODIS NDVI (Normalized Difference Vegetation Index) can explain the forest soil moisture simulated from SWAT (Soil and Water Assessment Tool) model. For ChungjuDam watershed ($6,661.3\;km^2$) which covers 82.2% of forest, the SWAT model was calibrated for four years (2003-2006) at two locations of the watershed using daily streamflow data and was verified for three years (2000-2002) with average Nash and Sutcliffe model efficiencies of 0.69 and 0.75 respectively. For the period from March to June, the average spatial correlation between 16 days composite MODIS NDVI and the corresponding SWAT forest soil moisture was 0.90. The two variables averaged for each data set during that period showed an inverse relation with the average coefficient of determination of 0.55.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.536-536
• /
• 2012

댐은 하천의 흐름을 막아 그 저수를 생활 및 공업용수, 농업용수, 발전, 홍수조절, 특정용도로 이용하기 위한 구조물을 일컫는다. 여러 가지 용도로 만든 댐의 저수지에는 상류에서 들어오는 유사가 저수지 바닥에 가라앉아 쌓이게 되는데 이를 저수지 퇴사(reservoir sedimentation)라 하며, 이는 저수지 유효 용량을 감소시키고 홍수시 유입 하천의 홍수위를 상승 및 저수지 수질을 악화시키는 등의 문제를 일으킨다. 저수지를 관리함에 있어 저수지로 유입되어 바닥에 가라앉는 퇴사량을 분석하고 저수용량과 수면적의 감소율을 예측하는 것은 효율적인 저수지 장기운영에서 매우 중요한 사항이다. 본 연구에서 향후 저수지로 유입되는 일유출량의 예측을 위해 과거 1987~2011년(25년)의 충주 댐 일유입량($m^3/s$) 자료를 단순반복시켜 향후 50년 동안의 일유량($m^3/s$)을 산정하였고, 일단위 모의가 가능한 유역단위의 분포형 장기 강우-유출모형인 SWAT를 이용하여 산정된 일유량($m^3/s$) 자료를 비교 평가하여 모형의 검증을 실시하였다. 유량-유사량 관계곡선을 이용하여 분석 대상 기간 동안의 총유사량을 구할 수 있으며, 한강유역조사(2002)에서 유도한 충주댐 상류 정선지점의 유량-유사량 관계식으로부터 향후 50년의 일유량($m^3/s$) 자료를 이용하여 총유사량를 산정하였다. 또한, 경험적 면적감소법을 이용하여 임의의 기간에 대한 실측치와 모의치 각각에 대한 퇴사분포 및 퇴사량을 산정 및 평가하였다. 이를 통해 효율적인 용수관리를 위한 저수지 퇴사관리 방안의 시기별 도출이 가능하였다.

• Journal of Korea Water Resources Association
• /
• v.36 no.1
• /
• pp.13-21
• /
• 2003

The purpose of this study is to estimate Z-R relationships of between radar reflectivity and rainfall rate. The Z-R relationships estimated that rainfall events are selected at Yeongchun water level station where the discharge recorded from 1,000cms to 8,519cms in chungju dam basin. The result of Z-R relationship distributed at thirty two raingage sites, the constant values of A and $\beta$ are distributed between 26.4 and 7.4, 0.9 and 1.56 respectively. The correlation coefficients of standard Z-R relationships(Z=200Rl.6)shows that 0.63 lower than each other raingage sites(0.65~0.748).