• Journal of The Korean Astronomical Society
• /
• v.34 no.1
• /
• pp.17-24
• /
• 2001

We have studied the central parts of M82, which is a well-known infrared luminous, starburst galaxy, by analyzing archival data from the Infrared Space Observatory (ISO). M82 was observed at 11 positions covering $\pm$45" from the center along the major axis. We analyzed 4 emission lines, [ArIII] 8.99 ${\mu}m$, $H_2$ 17.034 ${\mu}m$, [FeII] 25,98 ${\mu}m$, and [SiII] 34,815 ${\mu}m$ from $SWSO_2$ data. The integrated flux distributions of these lines are quite different. The $H_2$ line shows symmetric twin peaks at $\~$18" from the center, which is a general characteristic of molecular lines in starburst or barred galaxies. This line appears to be associated with the rotating molecular ring at around $\~$200 pc just outside the inner spiral arm. The relative depletion of the $H_2$ line at the center may be due to the active star formation activity which dissociates the $H_2$ molecules. The other lines have peaks at the center and the distributions are nearly symmetric. The line profiles are deconvolved assuming that both intrinsic and instrumental profiles are Gaussian. The velocity dispersion outside the core is found to be $\~50 km s^{-1}$. The central velocity dispersion is much higher than $50 km s^{-1}$, and different lines give different values. The large central velocity dispersion ($\sigma$) is mostly due to the rotation, but there is also evidence for a high $\sigma$ for [ArIII] line. We also generated position-velocity maps for these four lines. We found very diverse features from these maps.

• 한국해양학회지
• /
• v.7 no.2
• /
• pp.59-73
• /
• 1972

The heavy minerals of a barred estuarine and lagoonal sediments along the North Carolina coast have been studied with more than one hundred samples. Currents, salinity, and pH exhibit well-developed gradients from the upstream parts of the estuaries on the west toward the open ocean on the east. Twenty- four heavy minerals were identified in sediments of the study area. However, less than half of these occur frequently and the remainder exist only in minor quantities or trace amounts. Heavy minerals usually comprise less than 1% of the sample but vary from sample to sample. The maximum amount of heavy minerals in sediments of sounds and estuaries is 2.4% and in sediments of Outer Banks is 16.7%. Opaque minerals range from 10 to 85% of the total heavy mineral assemblage. Garnet and sillimanite are relatively more abundant in the eastern part than the western part in the area. Garnet more abundant in the northern part than the southern part, whereas sillimanite is more abundant in the southern part than the northern part, because the garnet source is in the northern part and one of the sillimanite source is in the southern part in the study area. The results of heavy mineral study indicate that the source of sediments is the Blue Ridge and Piedmont crystalline complex, and Coastal Plain formations. Some portions of sediments are transported from the Atlantic Ocean by the landward currents. They further indicate that the sediments of the Atlantic coast in the study area are transported mainly from the northern part to southern part by longshore littoral currents, and some portions of sediments are transported from the southern part to the northern part by the Gulf Stream.

• Journal of KIISE:Computer Systems and Theory
• /
• v.29 no.5
• /
• pp.257-264
• /
• 2002

In this paper, we give a detailed description of KDSM(KAIST Distributed Shared Memory) system. KDSM is implemented as a user-level library running on Linux 2.2.13, and TCP/IP is used for communication. KDSM uses page-based invalidation protocol, multiple-writer protocol, and supports HLRC(Home-based Lazy Release Consistency) memory consistency model. To evaluate performance of KDSM, we executed 4 scientific applications and compared the result to JLAJLA. The results showed that performance of KDSM almost equal to JIAJIA for 2 applications and performance of KDSM is better than JIAJIA for 2 applications.

• The Transactions of the Korean Institute of Electrical Engineers B
• /
• v.51 no.4
• /
• pp.210-220
• /
• 2002

We analysised the effect of harmonics on electric machines of substation power system barred on quantitatively measured harmonics and proposed the methods for prevention of harmonics through checking on transformer, rectifier and cable's capacities against harmonics with reference to KEPCO's electricity service standard. In order to analysis harmoninics of silicon rectifier that is power source in DC substation, computer simulations for a substation with TR of high voltage distribution switchboard are performed. Simulation results show that the total harmonic distortion factor becomes smaller for TR primary and receiving points in order rather than silicon rectifier which is harmonic generation source so that the harmonics generated frets each rectifier are outflowed to power supply and high voltage distribution switchboard The result of higher distortion factors of voltage and current for rectifier with 100% load than those with 50 % and 30% indicates that the waveform of voltage and current for the real substation power system at the office-going and the closing hours with heavy loads might be more distorted. As proposed methods for harmonic reduction, the conventional 6 pulse-type for substation is required to be replaced by 12 pulse-type for reduction of 5th and 7th harmonics. The active filter rather than the passive filter is more effective due to severe variance of rectifier loads, but the high cost is price to be paid. In view of installation area and costs, the use of 12 pulse-type transformer is desirable and then the parallel transformer and the rectifier within the substation must be replaced at the same time. Other substations with parallel feeder can use 6 pulse-type transformer.

• The Bulletin of The Korean Astronomical Society
• /
• v.35 no.1
• /
• pp.75.2-75.2
• /
• 2010

We investigate the dependence of occurrence of bar in galaxies on galaxy properties and environment. The environmental conditions considered include the large-scale background density and distance to the nearest neighbor galaxy. We use a volume-limited sample of 33,296 galaxies brighter than $M_r$=-19.5+5logh at $0.02{\leqq}z{\leqq}0.05489$, drawn from the Sloan Digital Sky Survey Data Release 7. We classify the galaxies into early and late types, and identify bars by visual inspection. We find that the fraction of barred galaxies ($f_{bar}$) is 18.2% on average in the case of late-type galaxies, and depends on both u-r color and central velocity dispersion $(\sigma);f_{bar}$ is a monotonically increasing function of u-r color, and has a maximum value at intermediate velocity dispersion (${\sigma}{\simeq}170km\;s^{-1}$). This trend suggests that bars are dominantly hosted by systems having intermediate-mass with no recent interaction or merger history. We also find that $f_{bar}$ does not directly depend on the large-scale background density as its dependence disappears when other physical parameters are fixed. We discover the bar fraction decreases as the separation to the nearest neighbor galaxy becomes smaller than 0.1 times the virial radius of the neighbor regardless of neighbor's morphology. These results imply that it is difficult for bars to be maintained during strong tidal interactions, and that the source for this phenomenon is gravitational and not hydrodynamical.

• The Bulletin of The Korean Astronomical Society
• /
• v.35 no.2
• /
• pp.33.1-33.1
• /
• 2010

We investigate the relation between the presence of bars in galaxies and AGN activities. Bars are believed to play an important role in fueling of AGN. Although there have been many previous studies on this topic, "the AGN-Bar Connection" is still an open question. To better understand the connection, we use a volume-limited sample of 9,726 late-type galaxies brighter than $M_r$=-19.5+5logh at $0.02{\leqq}z{\leqq}0.05489$, drawn from SDSS DR7. Among galaxies in the sample, 1,963 galaxies are classified as AGN-host galaxies based on the emission-line ratios while barred galaxies are identified by visual inspection. The bar fraction in AGN host galaxies (22.5%) is 3-times higher than in star-forming galaxies (8.6%). However, this trend is simply caused by the fact that the bar fraction increases with galaxy mass or luminosity and that AGN host galaxies are on average more massive than star-forming galaxies. Nevertheless, we find that among AGN host galaxies, the bar fraction increases with the Eddington ratio $(L_{[OIII]}/M_{[BH]})$, and this trend remains intact even at fixed galaxy luminosity and stellar velocity dispersion. These results imply that bars play a role in triggering AGNs.

• The Bulletin of The Korean Astronomical Society
• /
• v.42 no.1
• /
• pp.58.3-58.3
• /
• 2017

In a ${\Lambda}CDM$ universe, galaxies are believed to evolve by mergers and accretions. The debris resulting from such processes remains as diffuse, low-surface brightness structures, such as outer disks, stellar halos, and faint companions. These structures will give us fruitful insight into the recent mass assembly history of galaxies, but it is challenging to observe them due to their low surface brightness. In order to explore the structural properties of outskirts of nearby galaxies, we conduct deep wide-field imaging survey with KMTNet. Here we present a progress report of data reduction for the images of NGC 1291, a lenticular barred galaxy with outer rings. To achieve accurate flat fielding, we use dark sky flat and remove the sky gradient of each exposure with a polynomial fit. As a result, we are able to reach 1 sigma depth of ${\mu}_R{\sim}29.6\;mag\;arcsec^{-2}$. We expect to investigate the surface brightness profile of NGC 1291 in 1-D profile, and color-gradient in the outer part of the galaxy using the B- and R-band images.

• Geotechnical Engineering
• /
• v.13 no.5
• /
• pp.101-124
• /
• 1997

The root pile system is insitu soil reinforcement technique that uses a series of reticulately installed micropiles. In terms of mechanical improvement by means of grouted reinform ming elements, the root pile system is similar to the soil nailing system. The main difference between root piles and soil nailing are due to the fact that the reinforcing bars in root piles are normally grouted under high pressure and that the alignments of the reinforcing members differ. Recently, the root pile system has been broadly used to stabilize slopes and retain excavations. The accurate design of the root pile system is, however, a very difficult tass owing to geometric variety and statical indetermination, and to the difficulty in the soilfiles interaction analysis. As a result, moat of the current design methods have been heavily dependent on the experiences and approximate approach. This paper proposes a quasi-three dimensional method of analysis for the root pile system applied to the stabilization of slopes. The proposed methods of analysis include i) a technique to estimate the change in borehole radium as a function of the grout pressure as well as a function of the time when the grout pressure is applied, ii) a technique to evaluate quasi -three dimensional limit-equilibrium stability for sliding, iii) a technique to predict the stability with respect to plastic deformation of the soil between adjacent root piles, and iv) a quasi -three dimensional finite element technique to compute stresses and dis placements of the root pile structure barred on the generalized plane strain condition and composite unit cell concept talon형 with considerations of the group effect and knot effect. By using the proposed technique to estimate the change in borehole radius as a function of the grout pressure as well as a function of the time, the estimations are made and compar ed with the Kleyner 8l Krizek's experimental test results. Also by using the proposed quasi-three dimensional analytical method, analyses have been performed with the aim of pointing out the effects of various factors on the interaction behaviors of the root pile system.

• Journal of Biomedical Engineering Research
• /
• v.23 no.2
• /
• pp.87-99
• /
• 2002

This paper describes a method for overcoming the motion artifacts inherent in synthetic aperture(SA) imaging. based on the investigation results as to the influence of a target motion on synthetic aperture techniques. This method uses a region-based motion compensation approach in which only the axial motion is estimated and compensated for a given region of interest(ROI) under the assumption that the whole ROI moves uniformly The estimated axial motion is calculated with a crosscorrelation(CC) method at the Point where the focused signal has the maximum energy within the ROI. We also presents a method for estimating the axial motion using the autocorrelation(AC) method that is widely used to estimate average Doppler frequency Both computer simulations and in vivo experiments show that the proposed methods can improve greatly the spatial resolution and SNR of ultrasound imaging by implementing the SA techniques for two-way dynamic focusing without motion artifacts. In addition the AC-barred motion compensation method provides almost the same results as the CC-based one, but with a dramatically reduced computational complexity.

• The Korean Journal of Applied Statistics
• /
• v.15 no.1
• /
• pp.33-43
• /
• 2002

Various kinds of estimation methods have been developed for imputation of categorical missing data. They include category method, logistic regression, and association rule. In this study, we propose two fusions algorithms based on both neural network and voting scheme that combine the results of individual imputation methods. A Mont-Carlo simulation is used to compare the performance of these methods. Five factors used to simulate the missing data pattern are (1) input-output function, (2) data size, (3) noise of input-output function (4) proportion of missing data, and (5) pattern of missing data. Experimental study results indicate the following: when the data size is small and missing data proportion is large, modal category method, association rule, and neural network based fusion have better performances than the other methods. However, when the data size is small and correlation between input and missing output is strong, logistic regression and neural network barred fusion algorithm appear better than the others. When data size is large with low missing data proportion, a large noise, and strong correlation between input and missing output, neural networks based fusion algorithm turns out to be the best choice.