• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 하계학술대회 논문집 D
• /
• pp.2209-2211
• /
• 2004

Unlike the wavelet neural network, since a mother wavelet layer of the self-recurrent wavelet neural network (SRWNN) is composed of self-feedback neurons, it has the ability to store past information of the wavelet. Therefore we propose the prediction method for the nonlinear chaotic time series model using a SRWNN. The SRWNN model is learned for the modeling of a function such that the inputs arc known values of the time series and the output is the value in the future. The parameters of the network are tuned to minimize the difference between the nonlinear mapping of the chaotic time series and the output of SRWNN using the gradient-descent method for the adaptive backpropagation algorithm. Through the computer simulations, we demonstrate the feasibility and the effectiveness of our method for the prediction of the logistic map and the Mackey-Glass delay-differential equation as a nonlinear chaotic time series.

• 생태와환경
• /
• 제41권spc호
• /
• pp.1-10
• /
• 2008

We patterned water quality of agricultural reservoirs according to the differences of six physico-chemical environmental factors (TN, TP, DO, BOD, COD, and SS) using four different ordination methods: Principal Components Analysis (PCA), Detrended Correspondence Analysis (DCA), Nonmetric Multidimensional Scaling (NMS), and Isometric Feature Mapping (Isomap). The data set was obtained from the water quality monitoring networks operated by the Ministry of Agriculture and Forestry and the Ministry of Environments. Chlorophyll-${\alpha}$ displayed the highest correlation with COD, followed by TP, BOD, SS, and TN (p<0.01), while negatively correlated with altitude and bank height of the reservoirs (p<0.01). Although four different ordination methods similarly patterned the reservoirs according to the gradient of nutrient concentration, PCA and NMS appeared to be the most efficient methods to pattern water quality of reservoirs based on the explanation power. Considering variable scores in the ordination map, the concentration of nutrients was positively correlated with Chl-${\alpha}$, while negatively correlated with altitude and bank height. These ordination methods may help to pattern agricultural reservoirs according to their water quality characteristics.

• 건축역사연구
• /
• 제20권5호
• /
• pp.47-61
• /
• 2011

This study examined the background and change of building Yeongam eupseong and characteristics and construction method of its spatial structure. The results of this study can be summarized as follows. First, Yeongam eupseong was built for the same purpose of national border defense against the invasion of Japanese army as other Eupseongs were built and it was found first in the literature in the era of King Munjong of Joseon Dynasty with the scale of three gates and 4,369 cheok of castle wall. Such a scale was continued and in particular, according to flat structure, road network and gate position shown in the map, the form and road system of Yeongam eupseong before the 19th century were mostly identical with the current ones. Second, castle wall was built over foundation by using talcum after leveling the foundation soil and castle wall can be divided into outer wall, filling part and inner wall from the section. Outer wall was constructed by inserting and layering pebbles with big irregular stone, filling part was made with talcum and riprap and inner wall with soil in several layers by keeping gradient. This building method showed common features in inserting and layering pebbles for outer wall, use of protruded base and section form compared to Jeollado Eupseong.

• Steel and Composite Structures
• /
• 제47권6호
• /
• pp.759-779
• /
• 2023

The paper proposes two hybrid metaheuristic optimization and artificial neural network (ANN) methods for the close prediction of the ultimate axial compressive capacity of concentrically loaded concrete filled double skin steel tube (CFDST) columns. Two metaheuristic optimization, namely genetic algorithm (GA) and particle swarm optimization (PSO), approaches enable the dynamic training architecture underlying an ANN model by optimizing the number and sizes of hidden layers as well as the weights and biases of the neurons, simultaneously. The former is termed as GA-ANN, and the latter as PSO-ANN. These techniques utilize the gradient-based optimization with Bayesian regularization that enhances the optimization process. The proposed GA-ANN and PSO-ANN methods construct the predictive ANNs from 125 available experimental datasets and present the superior performance over standard ANNs. Both the hybrid GA-ANN and PSO-ANN methods are encoded within a user-friendly graphical interface that can reliably map out the accurate ultimate axial compressive capacity of CFDST columns with various geometry and material parameters.

• 환경위생공학
• /
• 제21권4호
• /
• pp.61-76
• /
• 2006

A Line Density algorithm was developed to quantify the sea surface temperature distribution using NOAA Sea Surface Temperature(SST) data and Geographic Information Systems(GIS), In addition, a GIS based automation model was designed to extract the Line Density Indices were determined by applying K-means Cluster. SST data in terms of March to May obtained on the coastal area of the Uljin from 2001 to 2004 in spring were used to make two data sets of average sea water temperature map in terms of year as well as month. From the result it was formed that water temperature gradient in April was the strongest among the other months, In particular very strog formation of oceanic front as well as temperature gradients were observed in front of the coastal area around Wonduk and Jukbyeon countries. Because those coastal area is a confront zone of two cold and a warm. It is expected that the development of a Line Density Algorithm would contribute to quantify of the SST for the research of Sea Surface Front(SSF) related to marine life management and the sea environmental conservation.

• 한국신재생에너지학회:학술대회논문집
• /
• 한국신재생에너지학회 2009년도 춘계학술대회 논문집
• /
• pp.600-603
• /
• 2009

We compared relative importance of thermal conductivity and initial ground temperature in designing borehole heat exchanger network and also we test accuracy of ground temperature estimation in thermal response test using a proven 3-D T-H modeler. The effect of error in estimating ground temperature on calculated total length of borehole heat exchanger was more than 3 times larger than the case of thermal conductivity in maximum 20% error range. Considering 10% of error in estimating thermal conductivity is generally acceptable, we have to define the initial ground temperature within 5% confidence level. Utilizing the mean annual ground surface temperature and the geothermal gradient map compiled so far can be a economic way of estimating ground temperature with some caution. When performing thermal response test for estimating ground temperature as well as measuring thermal conductivity, minimum 100 minutes of ambient circulation is required, which should be even more in case of very cold and hot seasons.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2004년도 춘계학술발표회
• /
• pp.219-226
• /
• 2004

We measure subsidences occurred in a reclaimed coastal land, Noksan industrial complex, by using JERS-1 SAR (1996-1998) and RADARSAT-1 SAR (2002-2003) dataset. SAR with a high spatial resolution (about several or several tens meter) can reveal the two-dimensional distribution of settlement that would be bardly estimated from in situ measurements. The DInSAR results show significant deformation signal associated with soil consolidation. Accuracy of the settlements estimated by 2-pass differential interferometry (DInSAR) is evaluated using the measurements of settlement gauge. A two-dimensional subsidence map is constructed from 7 qualified pairs. Comparing the JERS-1 radar measurements with the ground truth data yields the correlation coefficient of 0.87 (RMSE of 1.44 cm). The regression line shows the gradient of 1.04 and intercepts close to the origin, which implies that the unbiased settlement can be measured by DInSAR technique. The residual settlements are also detected from RADARSAT-1 pairs. The extent and amount of the settlements are matched well with ground truth data.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2005년도 학술발표회 논문집
• /
• pp.605-609
• /
• 2005

This study is to perform the rainfall-runoff analysis of the basin of Yongdam dam where is loacted in the Geumriver basin. The model used is the SAC-SMA model which was developed by U.S. National Weather Service. The Precipitation data used as the input data of the model are daily ones observed in 2002 and the mean of values recorded in 5 rainfall stations. The evaporation data are used observed in Daejeon meteorological station. The geographical data such as basin slope and stream gradient are elicited from the numerical map analysis. In the verification through the comparison of calculated daily inflow with observed one, parameters used in the model are estimated manually. As the result of verification, total annual calculated inflow is 13,547CMS and agree accurately with the observed one. During the period of one year of 2002, before 100 days and after 250 days, the soil moisture condition in the upper zone was significantly dry and in spite of the rainfall in this period, the runoff was not generated. Through this result, we can observe that the moisture condition in the soil affects strongly the runoff in a basin.

• 천문학회지
• /
• 제45권5호
• /
• pp.117-125
• /
• 2012

We present $^{12}CO$ J = 2-1 line observations of G54.1+0.3, a composite supernova remnant with a mid-infrared (MIR) loop surrounding the central pulsar wind nebula (PWN). We map an area of $12^{\prime}{\times}9^{\prime}$ around the PWN and its associated MIR loop. We confirm two velocity components that have been proposed to be possibly interacting with the PWN/MIR-loop; the +53 km $s^{-1}$ cloud, which appears in contact with the eastern boundary of the PWN and the +23 km $s^{-1}$ cloud, which has CO emission coincident with the MIR loop. However, we have not found a direct evidence for the interaction in either of these clouds. Instead, we detected an 5'-long arc-like cloud at +15-+23 km $s^{-1}$ with a systematic velocity gradient of ~3 km $s^{-1}$ $arcmin^{-1}$ and broad-line emitting CO gas with widths (FWHM) of ${\leq}7km\;s^{-1}$ in the western interior of the supernova remnant. We discuss their association with the supernova remnant.

• 한국정보통신학회논문지
• /
• 제10권10호
• /
• pp.1829-1834
• /
• 2006

물체의 경계로부터 거리에 따라 다른 크기의 창을 사용함으로써 물체의 경계를 정확히 찾는 새로운 스테레오정합 알고리즘을 제안한다. 원 영상을 같은 크기의 작은 부분으로 나누고, 이 작은 영역들의 중앙 화소들의 시차를 구한 표본 시차 지도로부터 축소된 물체의 경계를 찾는다. 경계로부터의 거리에 따라 크기가 다른 창을 원 영상에 적용하여 시차를 구한다. 원 영상의 시차를 구할 때 작은 영역 내에 있는 화소들의 시차는 이미 구한 중앙화소의 시차와 유사하다는 기울기 제한 조건을 이용하였다. 실험결과로부터 제안한 알고리즘이 다른 알고리즘 보다 우수함을 확인하였다.