• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2012.04a
• /
• pp.656-657
• /
• 2012

• Proceedings of the Korea Water Resources Association Conference
• /
• 1995.05a
• /
• pp.238-244
• /
• 1995

• Journal of the Korean Association of Geographic Information Studies
• /
• v.6 no.3
• /
• pp.129-138
• /
• 2003

A new methodology for selecting spatially variable model control parameter values through consideration of inference models within a Hydroinformatic system has been developed to overcome problems associated with determination of spatially variable control parameter values for both ungauged and gauged catchment. The adopted Hydroinformatic tools for determination of control parameter values were a GIS(Arc/Info) to handle spatial and non-spatial attribute information, the SWMM(stormwater management model) to simulate catchment response to hydrologic events, and lastly, L_BFGS_B(a limited memory quasi-Newton algorithm) to assist in the calibration process. As a result, high accuracy of control parameter estimation was obtained by considering the spatial variations of the control parameters based on landuse characteristics. Also, considerable time and effort necessary for estimating a large number of control parameters were reduced from the new calibration approach.

• Atmosphere
• /
• v.29 no.5
• /
• pp.567-583
• /
• 2019

In this study, the high-resolution numerical simulations considering detailed anthropogenic heat, albedo, emission and roughness length are analyzed by using single layer Urban Canopy Model (UCM) in Weather Research Forecast (WRF). For this, improved urban parameter data for Seoul Metropolitan Area (SMA) was collected from global data. And then the parameters were applied to WRF-UCM model after it was processed into 2-dimensional topographical data. The 6 experiments were simulated by using the model with each parameter and verified against observation from Automated Weather Station (AWS) and flux tower for the temperature and sensible heat flux. The data for sensible heat flux of flux towers on Jungnang and Bucheon, the temperature of AWS on Jungnang, Gangnam, Bucheon and Neonggok were used as verification data. In the case of summer, the improvement of simulation by using detailed anthropogenic heat was higher than the other experiments in sensible flux simulation. The results of winter case show improved in all simulations using each advanced parameters in temperature and sensible heat flux simulation. Improvement of urban parameters in this study are possible to reflect the heat characteristics of urban area. Especially, detailed application of anthropogenic heat contributed to the enhancement of predicted value for sensible heat flux and temperature.

• The Bulletin of The Korean Astronomical Society
• /
• v.41 no.1
• /
• pp.40.1-40.1
• /
• 2016

Recent integral-field spectrograph surveys have found that similar-looking early type galaxies have wide range of rotational properties (Emsellem et al. 2007). This finding initiated a new point of view to the galaxies; rotation of galaxy as the first parameter of galaxy classification (Emsellem et al. 2011, Cappellari et al. 2011, for example). Some theoretical studies tried to address the origin of galaxy rotation. Idealized galaxy merger simulations have shown that galaxy-galaxy interactions have significant effects on the rotation of galaxies. Cosmological simulations by Naab et al. 2014 also added some more insights to the rotation of galaxies. However, previous studies either lack cosmological background or have not enough number of samples. Running a set of cosmological hydrodynamic zoom-in simulations using the AMR code RAMSES(Teyssier 2002). we have constructed a sample of thousands of galaxies in 20 clusters. Here we present a kinematic analysis of a large sample of galaxies in the cosmological context. The overall distribution of rotation parameter of simulated galaxies suggests a single peak corresponding to fast rotating galaxies. But when divided by mass, we find a strong mass dependency of galaxy rotation, and massive galaxies are distinctively slow rotating. The cumulated effective of mergers seems to neutralize galaxy rotation as suggested by previous studies (Khochfar et al. 2011, Naab et al. 2014, and Moody et al. 2014). This is consistent with the fact that massive galaxies tend to rotate more slowly after numerous mergers. However, if seen individually, merger can either increase or decrease galaxy rotation depending on mass ratio, orbital parameter, and relative rotation axis of the two galaxies. This explains the existence of some non-slow rotating massive early type galaxies.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.49 no.2
• /
• pp.9-14
• /
• 2012

In this study, multi-finger RF MOSFET substrate parameters are accurately extracted by using S-parameters measured from common source-bulk and common source-gate test structures. Using this extraction method, the accuracy of an asymmetrical model with three substrate resistances is verified by observing better agreement with measured Y-parameters than a simple model with a single substrate resistance. The modeled S-parameters of the asymmetrical model also show excellent agreement with measured ones up to 20GHz.

• The Korean Journal of Applied Statistics
• /
• v.30 no.1
• /
• pp.69-82
• /
• 2017

R-indicator (representativeness indicator) is used to check the representativeness of samples when non-responses occur. The representativeness is related with the accuracy of parameter estimator and the accuracy is related with bias of the estimator. Hence, unbiased estimator generates high accuracy. Therefore, high value of R-indicator guarantees the accuracy of parameter estimation with a small bias. R-indicator is calculated through propensity scores obtained by logit or probit modeling. In this paper we investigate the degree of relation between R-indicator and different non-response rates in strata using simulation studies. We also analyze a modified Korea Economic Census data for real data analysis.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.451-452
• /
• 2023

최근 CNN 이 다양한 산업에 확산되고 있으며, IoT 기기 및 엣지 컴퓨팅에 적합한 경량 모델에 대한 연구가 급증하고 있다. 본 논문에서는 CNN 모델의 파라미터 비트 연산을 위한 자동화 프레임워크를 제안하고, 파라미터 비트와 모델 정확도 사이의 관계를 실험 및 연구한다. 제안된 프레임워크는 하위 n- bit 를 0 으로 설정하여 정보 손실 발생시킴으로써 ImageNet 데이터셋으로 사전 학습된 CNN 모델의 파라미터와 정확도의 강인성을 비트 단위로 체계적으로 실험할 수 있다. 우리는 비트 연산을 수행한 파라미터로 InceptionV3, InceptionResnetV2, ResNet50, Xception, DenseNet121, MobileNetV1, MobileNetV2 모델의 정확도를 평가한다. 실험 결과는 성능이 낮은 모델일수록 파라미터와 정확도 간의 강인성이 높아 성능이 좋은 모델보다 정확도를 유지하는 비트 수가 적다는 것을 보여준다.

• Journal of the Korea Safety Management & Science
• /
• v.10 no.2
• /
• pp.225-235
• /
• 2008

It is expected that the change of industrial structure of Knowledge based economy, and service based economy would affect on economic life of physically-challenged person with no exceptions. There have been some former studies which have been misunderstood the employment of disabled due to the declining employment of the disabled and prejudice to the disabled based on biased perspective and public opinion. This study have analyzed the employment conditions of the ordinary workers and disabled workers out of those prejudices and biased opinions. Wage, economic activity ratio, age, educational background and industry those 5 parameters are brought for analyzing the two comparison group. Each parameter have compared based on the statistical data of ordinary workers and disabled workers for past 7 years. Weight between each parameter have been drawn by collecting the experts' opinion using AHP(Analytic Hierarchy Process), and utility of employment of two comparison groups have been estimated through progress inclination of each parameter. Through the estimated outcome, present condition and controversial issues and way of improvement of domestic employment of ordinary workers and the disabled is mentioned.

• Korean Journal of Computational Design and Engineering
• /
• v.10 no.4
• /
• pp.284-292
• /
• 2005

This paper presents a new framework of kinematic tolerance synthesis and describes the implemented algorithm for planar mechanical systems comprised of higher kinematic pairs. Input to the synthesis algorithm is a parametric model of the mechanical system with allowed parameter ranges (tolerance ranges). The model is specified as the part profiles consisting of line and arc segments and the motion axes along which each part moves. The algorithm analyzes tolerance in generalized configuration space, called contact zones bounding the worst-case variations, and identifies bad system variations. The bad system variations then are removed out of the parameter ranges by adjusting the nominal parameter values if possible and then shrinking the ranges otherwise. This cycle is repeated until no more bad variations we found. I show the effectiveness of the algorithm by case studies on several mechanisms.