• Proceedings of the Korea Water Resources Association Conference
• /
• 2010.05a
• /
• pp.793-797
• /
• 2010

물리적 녹조저감 기술인 커튼형 수류차단막은 유입 하천과 저수지 천이부에서 높은 영양염류와 조류를 포함한 표층 수류의 차단 또는 우회를 통해 본류 수역의 녹조발생을 저감하는 대책이다. 본 연구에서는 2009년 5월 회남대교 약 2 km 상류에 시범 설치된 대청호의 수류차단막 효과를 분석하고자 선행연구에서 보정한 2차원 횡방향 평균 수리 및 수질 모델을 최근의 수문사상인 2009년 6~8월까지 적용하여 검보정하고 수치모의를 실시하였다. 저수지 수위와 실측수위를 비교한 결과, 7월 중순 유입량 증가에 따른 수위 상승을 잘 반영하였고, 결정계수값($R^2$)이 0.997로 나타나 모델은 저수지 물수지 계산에 있어서 높은 신뢰도를 보였다. 댐앞과 회남수역에서 수심별 수온예측 오차는 AME 0.258~$1.584^{\circ}C$, RMSE 0.393~$2.548^{\circ}C$의 범위로 실측값을 잘 반영하는 것으로 나타났다. 회남, 댐, 추동, 문의 수역의 표층에서 $PO_4$-P 및 Chl-a 농도에 대한 모의값과 실측값의 시계열 비교 결과, 모델은 저수지내 각 측정지점에서 실측값의 시계열 변화를 잘 모의하였고, 회남수역에서 7월 중순 홍수 유입에 따라 증가하였다. 특히, $PO_4$-P 농도가 0.06 mg/L까지 증가하는 것으로 나타났다. 이는 홍수기에 높은 영양염류를 포함한 탁수가 수류차단막 하단을 통과하여 수리학적 도수현상(Hydraulic jump)을 일으킨 것이 원인이라 판단된다. 수류차단막 설치에 따라 회남수역의 표층에서 Chl-a 농도의 저감 효과가 두드러졌으나, 댐, 추동 및 문의수역에서의 제어 효과는 미미하였다. 이와 같이 회남수역에서 효과가 큰 이유는 2009년 수문사상의 영향과 저수지 지형특성상 유입수의 영향을 직접받기 때문으로 판단된다. 또한, 회남수역에서 수류차단막 설치에 따른 T-N 및 T-P 평균 저감 효율(수류차단막이 설치되지 않은 경우에 대한 설치 후 농도 저감 비)은 각각 10.8% 및 19.1%이었으며, 평균 저감농도는 모의값을 기준으로 각각 1.637 mg/L에서 1.461 mg/L 및 0.047 mg/L에서 0.038 mg/L로 나타났다. DIN 및 $PO_4$-P 평균 저감 효율은 각각 6.4% 및 24.6%이었고, Chl-a 평균 저감 효율은 25.5%이었으며, 평균 저감농도는 모의값을 기준으로 0.025 mg/L에서 0.018 mg/L로 나타났다. 모의결과를 종합해 볼 때, 대청호에 시범 설치된 조류제어용 수류차단막은 2009년의 수문사상에서 회남수역의 녹조발생 저감에 기여한 것으로 판단된다. 또한 대청호에서 유사한 수문사상을 보인 2006년(62일간)에 비해 2009년(28일간)에 조류주의보 발령 일수가 대폭 줄었다는 사실도 차단막의 효과를 간접적으로 확인해 준다.

• Journal of the Korean Data and Information Science Society
• /
• v.21 no.3
• /
• pp.419-425
• /
• 2010

Kernel machine learning is gaining a lot of popularities in analyzing large or high dimensional nonlinear data. We use this technique to estimate a GARCH model for predicting the conditional volatility of stock market returns. GARCH models are usually estimated using maximum likelihood (ML) procedures, assuming that the data are normally distributed. In this paper, we show that GARCH models can be estimated using kernel machine learning and that kernel machine has a higher predicting ability than ML methods and support vector machine, when estimating volatility of financial time series data with fat tail.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.11 no.3
• /
• pp.81-91
• /
• 2012

The world has been replacing fast fossil fuels vehicles with electric vehicles(EVs) to cope with climate change. The government set a goal which EVs will be substitute at least 10% of the domestic small vehicles with EVs until 2020, and will try to build electric charging infrastructures in apartments with the revision the law of 'the housing construction standards'. In apartments the EVs charging infrastructure and parking space is, essential to accomplish the goal. But the studies on EVs demand are few. In this study, we predicted that the demand for EVs using time-series analysis of statistical data, survey results for apartments residents in the metropolitan area. As a result, the ratio of the EVs appeared to be 6~21% for the total vehicles in a rental apartments for the years 2020, 21~39% in apartments for sales. For the EVs, the maximum power required for 1,000 households in rental apartment is predicted to be about 4200 kwh on a daily basis, while the maximum power in the apartment for sales is predicted to be 7800kwh.

• Journal of Korea Port Economic Association
• /
• v.26 no.2
• /
• pp.49-57
• /
• 2010

This paper examines the relationship between port import from China and macroeconomic variables such as international financial crisis, exchange rate, and industrial production during the period 2000-2009. I employ GPH cointegration methodology since the model must be stationary to avoid the spurious results. The empirical results show that our model is stationary as well as mean-reverting. This paper also applies impulse-response functions to get additional information regarding the responses of the port import to the shocks economic variables such as financial crisis, exchange rate, and industrial production. The results show that the response of port import to exchange rate and financial crisis declines at the first and dies out slowly.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.30 no.6B
• /
• pp.561-569
• /
• 2010

In this study, the impact of climate change on extreme drought events is investigated by comparing drought severity-area-duration curves under present and future climate. The depth-area-duration analysis for characterizing an extreme precipitation event provides a basis for analysing drought events when storm depth is replaced by an appropriate measure of drought severity. In our climate-change impact experiments, the future monthly precipitation time series is based on a KMA regional climate model which has a $27km{\times}27km$ spatial resolution, and the drought severity is computed using the standardized precipitation index. As a result, agricultural drought risk is likely to increase especially in short duration, while hydrologic drought risk will greatly increase in all durations. Such results indicate that a climate change vulnerability assessment for present water resources supply system is urgent.

• KDI Journal of Economic Policy
• /
• v.19 no.4
• /
• pp.3-46
• /
• 1997

본 논문은 우리나라 저축률(貯蓄率)의 결정요인을 실증적으로 분석하는 데 목적을 두고 있다. 특히 본 논문은 생애주기가설/항상소득가설(生涯週期假說/恒常所得假說)에 바탕을 두고 거시(巨視)시계열자료와 미시(微視)횡단면자료를 모두 분석함으로써 개별경제주체의 저축행위와 경제전체의 저축간의 일관된 관계를 밝히려고 하였다. 표준적인 생애주기가설/항상소득가설에 의하면, 저축은 소득(所得)의 일시적 변동으로부터 소비(消費)를 독립시키려는 개별소비자의 합리적 선택의 결과이다. 따라서 개별소비자의 저축은 단기적으로는 소득이 일시적으로 높은 해(년(年))에, 그리고 보다 장기적으로는 일생동안 가장 높은 수준의 소득을 벌게 되는 장년기간중에 가장 크게 된다. 본 논문의 실증결과는 이러한 생애주기가설/항상소득가설의 예측이 실제자료와 대체로 일치함을 보여준다. 거시자료 분석결과에 의하면 우리나라 저축률의 연간변동은 소득성장률(所得成長率)과 인구연령구조(人口年齡構造)의 변동에 의해서 잘 설명되는 것으로 나타난다. 또한 미시자료 분석결과를 보더라도 소득이 일시적으로 높은 가계나 경제활동연령인구의 비중이 높은 가계일수록 더 많은 저축을 하는 것으로 나타난다. 따라서 생애주기가설/항상소득가설은 우리나라 저축률의 결정을 설명하는 데 매우 유용하다고 판단된다. 본 논문은 또한 소득성장률이나 연령구조 외에 이자율, 유동성 제약, 그리고 예비적 저축동기 등이 저축에 미치는 영향에 대해서도 살펴보았다. 실증결과에 의하면 실질이자율은 저축률을 다소 증대시키는 효과가 있으나, 기타 요인들의 효과는 유의하지 않은 것으로 나타났다.

• Journal of Korea Water Resources Association
• /
• v.53 no.4
• /
• pp.273-283
• /
• 2020

Quantitative forecasting of groundwater levels for the assessment of groundwater variation and vulnerability is very important. To achieve this purpose, various time series analysis and machine learning techniques have been used. In this study, we developed a prediction model based on LSTM (Long short term memory), one of the artificial neural network (ANN) algorithms, for predicting the daily groundwater level of 11 groundwater wells in Hankyung-myeon, Jeju Island. In general, the groundwater level in Jeju Island is highly autocorrelated with tides and reflected the effects of precipitation. In order to construct an input and output variables based on the characteristics of addressing data, the precipitation data of the corresponding period was added to the groundwater level data. The LSTM neural network was trained using the initial 365-day data showing the four seasons and the remaining data were used for verification to evaluate the fitness of the predictive model. The model was developed using Keras, a Python-based deep learning framework, and the NVIDIA CUDA architecture was implemented to enhance the learning speed. As a result of learning and verifying the groundwater level variation using the LSTM neural network, the coefficient of determination (R²) was 0.98 on average, indicating that the predictive model developed was very accurate.

• Journal of Digital Convergence
• /
• v.18 no.11
• /
• pp.259-266
• /
• 2020

Many developments have been steadily carried out by researchers with applying knowledge-based expert system or machine learning algorithms to the financial field. In particular, it is now common to perform knowledge based system trading in using stock prices. Recently, deep learning technologies have been applied to real fields of stock trading marketplace as GPU performance and large scaled data have been supported enough. Especially, LSTM has been tried to apply to stock price prediction because of its compatibility for time series data. In this paper, we implement stock price prediction using LSTM. In modeling of LSTM, we propose a fitness combination of model parameters and activation functions for best performance. Specifically, we propose suitable selection methods of initializers of weights and bias, regularizers to avoid over-fitting, activation functions and optimization methods. We also compare model performances according to the different selections of the above important modeling considering factors on the real-world stock price data of global major companies. Finally, our experimental work brings a fitness method of applying LSTM model to stock price prediction.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.27 no.1
• /
• pp.22-28
• /
• 2021

This is a basic study on the development of deep learning-based algorithms to detect smoke before the smoke detector operates in the event of a ship fire, analyze and utilize the detected data, and support fire suppression and evacuation activities by predicting the spread of smoke before it spreads to remote areas. Proposed algorithms were reviewed in accordance with the following procedures. As a first step, smoke images obtained through fire simulation were applied to the YOLO (You Only Look Once) model, which is a deep learning-based object detection algorithm. The mean average precision (mAP) of the trained YOLO model was measured to be 98.71%, and smoke was detected at a processing speed of 9 frames per second (FPS). The second step was to estimate the spread of smoke using the coordinates of the boundary box, from which was utilized to extract the smoke geometry from YOLO. This smoke geometry was then applied to the time series prediction algorithm, long short-term memory (LSTM). As a result, smoke spread data obtained from the coordinates of the boundary box between the estimated fire occurrence and 30 s were entered into the LSTM learning model to predict smoke spread data from 31 s to 90 s in the smoke image of a fast fire obtained from fire simulation. The average square root error between the estimated spread of smoke and its predicted value was 2.74.

• Journal of Korea Water Resources Association
• /
• v.31 no.3
• /
• pp.317-326
• /
• 1998

Water consumption characteristics on the northern part of Seoul were analyzed using wavelet transform with a base function of Coiflets 5. It turns out that long term evolution mode detected at 212 scale in 1995 was in a shape of hyperbolic tangent over the entire period due to the development of Sanggae resident site. Furthermore, there was seasonal water demand having something to do with economic cycle which reached its peak at the ends of June and December. The amount of this additional consumption was about $1,700\;\textrm{cm}^3/hr$ on June and $500\;\textrm{cm}^3/hr$ on December. It was also shown that the periods of energy containing sinusoidal component were 3.13 day, 33.33 hr, 23.98 hr and 12 hr, respectively, and the amplitude of 23.98 hr component was the most humongous. The components of relatively short frequency detected at $2^i$[i = 1,2,…12] scale were following Gaussian PDF. The most reliable predictive models are multiple AR[32,16,23] and ARMA[20, 16, 10, 23] which the input of temperature from the view point of minimized predictive error, mutual independence or residuals and the availableness of reliable meteorological data. The predicted values of water supply were quite consistent with the measured data which cast a possibility of the deployment of the predictive model developed in this study for the optimal management of water supply facilities.