• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.31 no.2B
• /
• pp.155-163
• /
• 2011

In this study, a soil moisture estimation model was developed using a decision tree model, an artificial neural networks (ANN) model, remotely sensed data, and ground network data of daily precipitation, soil moisture and surface temperature. Soil moisture data of the Yongdam dam basin (5 sites) were used for model validation. Satellite remote sensing data and geographical data and meteorological data were used in the classification and regression tree (CART) model for data classification and the ANNs model was applied for clustered data to estimate soil moisture. Soil moisture data of Jucheon, Bugui, Sangjeon, Ahncheon sites were used for training and the correlation coefficient between soil moisture estimates and observations was between 0.92 to 0.96, root mean square error was between 1.00 to 1.88%, and mean absolute error was between 0.75 to 1.45%. Cheoncheon2 site was used for validation. Test statistics showed that the correlation coefficient, the root mean square error, the mean absolute error were 0.91, 3.19%, and 2.72% respectively. Results demonstrated that the developed soil moisture model using CART and ANN was able to apply for the estimation of soil moisture distribution.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.399-400
• /
• 2017

홍수나 가뭄 등 극치 현상의 통계분석 및 빈도해석에 있어 극치분포형이 널리 사용되고 있으며, 이러한 극치분포형의 특성을 이해하기 위해서는 분포형의 오른쪽 꼬리(right tail) 부분 특성을 자세히 분석할 필요가 있다. 이에 따라 본 연구에서는 Monte Carlo 모의를 통하여 다양한 극치분포형의 오른쪽 꼬리 부분의 통계적 특성 및 그 예측 능력을 연구하였다. 극치분포형으로는 우리나라 확률수문량 산정에 널리 활용되고 있는 generalized extreme value (GEV), Gumbel, generalized logistic 분포를 사용하였으며, 매개변수 산정 방법으로는 확률가중모멘트법을 사용하였다. 모의실험의 모분포로는 수문빈도해석에서 많이 사용되는 GEV 분포를 사용하였고, 30년 이상 자료를 보유한 기상청 지점 자료의 왜곡도를 조사하여 모의실험에 사용되는 모집단의 왜곡도로 가정하여 표본 자료를 발생시켰다. 예측 능력의 평가는 재현기간 10~1000년의 확률수문량을 왜곡도계수를 고려한 GEV 도시위치공식을 이용하여 GEV 확률지에 도시하고, 평균제곱근오차(root mean square error), 편의(bias), 평균상대오차(mean relative difference), 평균절대상대오차(mean absolute relative difference)를 이용하여 최적 분포형을 선정함으로써 이루어진다. 또한 예측 능력 평가결과의 타당성 확인을 위해 극치분포형의 적합정도를 잘 나타낸다고 알려진 modified Anderson-Darling 방법의 검정결과와 비교하여 적절성을 확인하였다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.9 no.5
• /
• pp.521-529
• /
• 2003

In this paper we present an improved method by using demographic information for overcoming the similarity miss-calculation from the sparsity problem in collaborative filtering recommendation systems. The similarity between a pair of users is only determined by the ratings given to co-rated items, so items that have not been rated by both users are ignored. To solve this problem, we add virtual neighbor's rating using demographic information of neighbors for improving prediction accuracy. It is one kind of extentions of traditional collaborative filtering methods using the peason correlation coefficient. We used the Grouplens movie rating data in experiment and we have compared the proposed method with the collaborative filtering methods by the mean absolute error and receive operating characteristic values. The results show that the proposed method is more efficient than the collaborative filtering methods using the pearson correlation coefficient about 9% in MAE and 13% in sensitivity of ROC.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2006.05a
• /
• pp.1822-1829
• /
• 2006

본 연구에서는 계산된 유량과 실측 유량을 비교하여 Clark 단위도 방법의 매개변수를 추정하고자 하였다. 오산천과 진위천 상류유역에 대하여 Arcview와 WMS로 지형자료에 대한 전 처리를 한후, HEC-HMS 프로그램을 이용하여 유출량을 산정하였다. 2001년부터 2005년까지 4개의 사상에 대하여 강우량, 기흥저수지와 이동저수지의 실제 방류량을 이용하여 유출량을 산정하였으며, Clark 모형의 매개변수를 Russel 공식, Sabol 공식 및 HEC-HMS 프로그램에 내장된 Nelder-Mead 최적화 방법을 이용하여 매개변수를 각각 산정하여 회화 지점의 실측 유출량과 비교.평가하였다. 빈도가 큰 유출사상의 경우에는 Sabol 식을 적용한 결과가 Russel 식을 적용한 모의결과보다 첨두유량의 재현성이 우수하게 나타났으며, 유출량이 작은 경우에는 Russel 식을 적용한 모의결과가 우수하였다. 첨두가 중제곱평균제곱근오차, 잔차자승의 합, 절대잔차의 합 등 3가지의 서로다른 목적함수를 적용하여 매개변수를 자동 보정하였을 때, 목적함수에 따른 첨두유량의 오차는 거의 동일하였으며, 첨두시간에 대한 오차는 첨두가중제곱평균제곱근오차를 적용했을 때 가장 작은 것으로 분석되었다. 그리고 Clark 유역 추적모형의 자동보정을 통하여 추정한 매개변수인 도달시간과 저류상수는 강우사상에 따라서 변동하는 특성을 나타내기 때문에 최적의 도달시간 및 저류상수는 홍수사상별로 추정되어야 하며 이 결과는 홍수량 산정을 위한 매개변수 추정과정의 비유일성 및 복잡성을 암시하고 있다.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.887-888
• /
• 2008

본 논문은 여러 음향신호의 도달각에 기반한 실내측 위성능을 분석한다. 도달각 측정을 위해 송출전력이 같은 총성, 벨소리, 그리고 의사충격음을 사용하였다. 음향 신호는 약 32 평방미터의 실험실에서 한 개의 음원 스피커를 통해 송출되고, 두 쌍의 서로 동기가 맞는 배열 마이크를 통해 수신된다. 수신된 음향신호로부터 한 마이크 쌍 사이의 도달지연시간차가 측정되고, 이로부터 도달각이 산출된다. 끝으로, 음원 스피커의 최종 위치는 도달각의 교점으로 결정된다. 본 논문은 설치한 실험환경에서의 참 도달각과 측정 도달각의 평균절대오차와 측위오차의 상보누적분포를 보이고, 이에 대한 성능특성을 분석한다.

• Fire Science and Engineering
• /
• v.30 no.6
• /
• pp.31-36
• /
• 2016

The flash point is one of the most important properties for characterizing the fire and explosion hazard of liquid solutions. In this study, the flash points of two flammable binary mixtures, n-pentanol + n-propanol and n-pentanol + n-heptanol systems were measured using a Seta flash closed cup tester. The flash point was estimated using the methods based on Raoult's law and multiple regression analysis. The measured flash points were also compared with the predicted flash points. The absolute average errors (AAE) of the results calculated by Raout's law were $1.3^{\circ}C$ and $1.3^{\circ}C$ for the n-pentanol + n-propanol and n-pentanol + n-heptanol mixtures, respectively. The absolute average errors of the results calculated by multiple regression analysis were $0.4^{\circ}C$ and $0.3^{\circ}C$ for the n-pentanol + n-propanol and n-pentanol + n-heptanol mixtures, respectively. According to the AAE, the calculated values based on multiple regression analysis were better than those based on Raoult's law.

• Korean Chemical Engineering Research
• /
• v.60 no.3
• /
• pp.363-370
• /
• 2022

In this paper, an estimation algorithm for state of charge (SOC) was applied using an equivalent circuit model (ECM) and an Extended Kalman Filter (EKF) to improve the estimation accuracy of the battery system states. In particular, an observer was designed to estimate SOC along with the aged capacity. In the case of the fresh battery, when SOC was estimated by Kalman Filter (KF), the mean absolute percentage error (MAPE) was 0.27% which was smaller than MAPE of 1.43% when the SOC was calculated by the model without the observer. In the driving mode of the vehicle, the general KF or EKF algorithm cannot be used to estimate both SOC and capacity. Considering that the battery aging does not occur in a short period of time, a strategy of periodically estimating the battery capacity during charging was proposed. In the charging mode, since the current is fixed at some intervals, a strategy for estimating the capacity along with the SOC in this situation was suggested. When the current was fixed, MAPE of SOC estimation was 0.54%, and the MAPE of capacity estimation was 2.24%. Since the current is fixed when charging, it is feasible to estimate the battery capacity and SOC simultaneously using the general EKF. This method can be used to periodically perform battery capacity correction when charging the battery. When driving, the SOC can be estimated using EKF with the corrected capacity.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.8C
• /
• pp.832-837
• /
• 2005

This paper presents a design method for fixed-width squarer that receives a W-bit input and produces a W-bit squared product. To efficiently compensate for the quantization error, modified Booth encoder signals (not multiplier coefficients) are used for the generation of error compensation bias. The truncated bits are divided into two groups (major/minor group) depending upon their effects on the quantization error. Then, different error compensation methods are applied to each group. By simulations, it is shown that the performance of the proposed method is close to that of the rounding method and much better than that of the truncation method and conventional method. It is also shown that the proposed method leads to up to $28\%\;and\;27\%$ reduction in area and power consumption compared with the ideal squarers, respectively.

• Analytical Science and Technology
• /
• v.30 no.6
• /
• pp.405-410
• /
• 2017

High energetic materials (HEMs) have been used in fuels, civil engineering and architecture as well as military purposes such as explosives and propellants. The essential process for the development of new energetic compounds is to accurately calculate its detonation performances. The most typical equation for calculating the explosive performance is the Kamlet-Jacobs (K-J) equation. In the K-J equation, the parameter such as the number of moles of gaseous products at the explosion, the average molar mass of gas products, and the explosion heat greatly affect the explosion performance. These depend on the product composition for the detonation reaction. In this study, detonation products of 65 high energetic molecules (HEMs) were calculated from the various rules such as Kamlet-Jacobs, Kistiakowsky-Wilson, modified Kistiakowsky-Wilson, Springall-Roberts rules to calculate more accurate detonation velocity (Dv). In addition, they were applied to five kinds of detonation velocity equations proposed by K-J, Rothstein, Xiong, Stine and Keshavarz. The mean absolute error and root mean square error of HEMs were obtained from experimental and calculated velocity value for each method. The K-J and Xiong equation that is slightly complex showed a lower mean absolute error than the simple Rothstein and Keshavarz equation. When the mod-KW rule was applied to the Xiong equation, the detonation velocities were the most accurate. This study compared the various method of calculating the detonation velocity of HEMs to obtain accurate HEMs performance.

• 한국IT서비스학회:학술대회논문집
• /
• 2009.11a
• /
• pp.491-496
• /
• 2009

우리의 생활 속 GPS를 이용한 서비스의 활용분야는 점점 더 넓어지고 있다. 사용자들이 요구하는 GPS 서비스 품질이 높아지면서 GPS의 정확성을 저하시키는 제한적인 상황들은 정확성 제고를 위한 보정을 필요로 한다. 본 연구에서는 도심 내 단층건물 밀집지역이 GPS 정확성을 떨어뜨리는 오차 요인이라고 가정하고 실험을 수행하였다. 평균 2m 간격으로 밀집된 단층 건물들 사이에서의 GPS의 정확성을 측정하였으며 그 결과 밀집된 단층건물에서 다중경로 효과로 인한 위도 경도 상의 오차범위와 절대적 오차범위를 산출할 수 있었다. 이것을 바탕으로 단층건물의 밀집도가 관측점에 위치되어 있는 고층건물보다는 GPS 수신에 더 적은 영향을 미친다는 결론을 도출하였다. 이러한 GPS 정확성은 앞으로 밀집도가 서로 다른 지역에서의 GPS 서비스와 건물의 높이에 따른 GPS 정확성 연구에 활용될 수 있을 것으로 사료된다.