• 응용통계연구
• /
• 제10권1호
• /
• pp.69-84
• /
• 1997

선형 캘리브레이션 실험계획 문제에 대하여, 베이지안 의사결정론을 이용하여 평균제곱오차손실을 최소화한 Kim(1988, 1993)의 실험계획과 관련 문헌의 결과인 몇 가지 최적계획을 비교한다. 비교대상 실험계획으로서 고전적 추정량의 점근분산을 최소화하는 Buonaccorsi(1986)의 최적계획, 회귀분석 모형에서 $ M(x) = \sum x_i x_i '$의 함수를 최대화 또는 최소화하는 D-optimal 또는 A-optimal 계획, Hunter and Lamboy(1981)가 베이지안 추정량의 특성을 설명하기 위하여 그 논문에서 예로 들었던 실험계획을 고려한다. 서로 다른 기준에 의한 최적계획을 비교하기 위해서 우선 기대사후분산을 계산하여 비교하고 몇가지 사전분포에 대하여 몬테칼로 시뮬레이션을 통한 평균분산과 HPD 구간의 크기를 비교한다.

• 한국수자원학회논문집
• /
• 제32권6호
• /
• pp.665-674
• /
• 1999

본 논문은 LRCS 모형을 낙동강 수계 송리원 수위표 지점의 18개 호우사상에 대하여 보정 및 민감도 분석을 실시하여 모형의 국내 하천에 적용성을 확인하였고, 모형의 적합성 분석 결과 LAD 추정자가 LS와 WLS에 비하여 좋지 못한 결과를 얻었다. "hot" 행렬의 대각 요소(hi)와 영향성 척도(Di)를 매개변수 추정치를 분석하는데 사용할 수 있고, 매개변수 IL이 모형 출력에 가장 중요함을 알 수 있었다. 매개변수의 오차에 따른 오차 전파의 정도는 IL, TP, F1 의 순서이며, 이는 모형의 출력이 첨두유량일 경우에 민감도의 정도를 나타낸다. 민감도 계수와 hi 의 대각 요소와 Di 값의 분석과 모형의 보정 및 민감도 분석의 관련성을 확인하였다.련성을 확인하였다.

• 한국염색가공학회지
• /
• 제32권4호
• /
• pp.217-225
• /
• 2020

In this study, the dyeing characteristics of nylon fabric which is dyed with supercritical fluid were investigated. There were two dyes used in the dyeing experiment: C.I. Disperse Red 167 and C.I. Disperse Violet 93. Dyeing temperature, pressure, and leveling time were fixed at 110℃, 250bar, 60minutes, and the experiment was conducted with dyeing concentration of 0.1, 0.3, 0.5, and 0.85% o.w.f. The analysis of the experimental results was found out through the measurement of washing fastness and color coordinate. In addition, the calibration curve of each dye was drawn up and the amount of remaining dye was checked by measuring the absorbance of the residual dye. As a result of color difference measurement, as the concentration increased, the L⁎ value decreased and the K/S value increased. However, the increase in K/S value compared to the amount of input decreased as the concentration increased. The comparative experiment on the amount of residual dye(C.I. Disperse Red 167) in the pot showed that 99.14% of the amount was dyed at the concentration of 0.1% o.w.f, while it rapidly decreased to 77% at 0.85% o.w.f. C.I. Disperse Violet 93 dye also decreased from 0.5% o.w.f to 93.91%. In the washing fastness experiment of both dyes, the level of washing fastness began to decrease from samples dyed at 0.5% o.w.f. It may be because the simply absorbed dye was produced instead of completely being fixed in the amorphous region of the nylon fiber.

• Nuclear Engineering and Technology
• /
• 제53권10호
• /
• pp.3421-3430
• /
• 2021

Spherical NaI(Tl) detectors are used in gamma-ray spectrometry, where the gamma emissions come from the nuclei with energies in the range from a few keV up to 10 MeV. A spherical detector is aimed to give a good response to photons, which depends on their direction of travel concerning the detector center. Some distortions in the response of a gamma-ray detector with a different geometry can occur because of the non-uniform position of the source from the detector surface. The present work describes the calibration of a NaI(Tl) spherical detector using both an experimental technique and a numerical simulation method (NSM). The NSM is based on an efficiency transfer method (ETM, calculating the effective solid angle, the total efficiency, and the full-energy peak efficiency). Besides, there is a high probability for a source-to-detector distance less than 15 cm to have pulse coincidence summing (CS), which may occur when two successive photons of different energies from the same source are detected within a very short response time. Therefore, γ-γ ray CS factors are calculated numerically for a ¹⁵²Eu radioactive cylindrical source. The CS factors obtained are applied to correct the measured efficiency values for the radioactive volumetric source at different energies. The results show a good agreement between the NSM and the experimental values (after correction with the CS factors).

• 한국염색가공학회지
• /
• 제34권2호
• /
• pp.93-101
• /
• 2022

In this study, the dyeing properties of supercritical fluid dyed cotton fabrics were investigated which use two types of dyes, such as C.I. Disperse orange and C.I. Disperse red 167. Dyeing temperature, pressure and leveling time were equally applied at 130 ℃, 250 bar, and 60 minutes with reference to the related literature, and experiments were performed at concentrations of 0.04, 0.1, 0.4 and 0.8 % o.w.f with different concentrations. Dyeability was confirmed through measurement of washing fastness and color coordinate, and a calibration curve of each dye was drawn up and the absorbance of the residual dye was measured to confirm the amount of residual dye and the dye exhaustion rate at the corresponding concentration. As a result of color difference measurement, as the concentration increased, the L^* value decreased and the K/S value increased. However, as the concentration increased, the increase in K/S value decreased compared to the input amount, and this tendency was more obvious in C.I. Disperse red 167 than in C.I. Disperse orange 155. The dye exhaustion rate which was calculated by using the amount of residual dye in the pot was also C.I. Disperse orange 155 was 96.16 % and C.I. Disperse red 167 was 94.57 %. However, as the dyeing concentration increased, the dye exhaustion rate decreased, that C.I. Disperse orange was 95.33 % and C.I. Disperse red 167 was 90.63 %. As a result of the washing fastness test for both dyes, dyed samples of which concentrations were 0.4 and 0.8 % o.w.f decreased by 0.5 ~ 1.0 grade. This is predicted because the dye did not completely adhere to the amorphous region of the cotton fiber and the dye simply adsorbed. The fastness to rubbing also maintained at least grade 3-4 up to the 0.1 % o.w.f concentration, but at the concentration of 0.4 % o.w.f or higher, it fell to grade 1 or lower, showing a very poor friction fastness.

• 천문학회지
• /
• 제17권2호
• /
• pp.69-103
• /
• 1984

Five different calibrations of metal abundances of globular clusters are examined and these are compared with metallicity ranking parameters such as $(Sp)_c$, . Q39 and IR-indices. Except for the calibration $[Fe/H]_H$ by the high dispersion echelle analysis. the other calibration scales are correlated with the morphological parameters of red giant branch. In the $[Fe/H]_H$-scale. the clusters later than ${\sim}F8$ have nearly a constant metal abundance. $[Fe/H]_H{\simeq}-1.05$, regradless of morphological characteristics of horizontal branch and red giant branch. By the two fundamental calibration scales of $[Fe/H]_L$ (derived by the low dispersion analysis) and $[Fe/H]_{{\Delta}s}$ (derived by the spectral analysis of RR Lyrae stars). the globular clusters are divided into the halo clusters with [Fe/H]<-1.0 and the disk clusters confined within the galactocentric distance ${\tau}_G=10\;kpc$ and galactic plane distance |z|=3 kpc. In this case the abundance gradient is given by d[Fe/H]/$dr_G{\approx}-0.05\;kpc^{-1}$ and d[Fe/H]/$d|z|{\simeq}-0.08\;kpc^{-1}$ within ${\tau}_G=20\;kpc$ and |z|=10 kpc, respectively. According to these characteristics of the spatial distribution of globular clusters. the chemical evolution of the galactic globular clusters can be accounted for by the two-zone (disk-halo) slow collapse model when the $[Fe/H]_L$-or $[Fe/H]_{{\Delta}s}$-scale is applied. In the case of $[Fe/H]_H$-scale, the one-zone fast collapse model is preferred for the evolution of globular clusters.