• Communications for Statistical Applications and Methods
• /
• v.29 no.3
• /
• pp.287-299
• /
• 2022

In radiation epidemiology, the excess relative risk (ERR) model is used to determine the dose-response relationship. In general, the dose-response relationship for the ERR model is assumed to be linear, linear-quadratic, linear-threshold, quadratic, and so on. However, since none of these functions dominate other functions for expressing the dose-response relationship, a Bayesian semiparametric method using splines has recently been proposed. Thus, we improve the Bayesian semiparametric method for the selection of the tuning parameters for splines as the number and location of knots using a Bayesian knot selection method. Equally spaced knots cannot capture the characteristic of radiation exposed dose distribution which is highly skewed in general. Therefore, we propose a nonparametric Bayesian knot selection method based on a Dirichlet process mixture model. Inference of the spline coefficients after obtaining the number and location of knots is performed in the Bayesian framework. We apply this approach to the life span study cohort data from the radiation effects research foundation in Japan, and the results illustrate that the proposed method provides competitive curve estimates for the dose-response curve and relatively stable credible intervals for the curve.

• Journal of the Korean Geotechnical Society
• /
• v.19 no.4
• /
• pp.299-310
• /
• 2003

In the evaluation of the subgrade stiffness structure by the SASW method, the calculation of the phase velocities is the important task controlling the reliability of the result. The interpretation of the phase spectrum should precede the phase-velocity calculation in the current practice of the SASW method. The difficulty involved in the interpretation prohibited the SASW method from being spread over to the industry. This study proposed a new method called the frequency-wave number technique, which is based on the frequency-wave number relationship of the surface wave in the multi-layered system. The frequency-wave number technique eliminates the expertise in the interpretation of the phase spectrum, automates the phase-velocity calculation and expedites the determination of the phase-velocity dispersion curve. To verify the validity of the proposed frequency-wave number method, the transfer function determined from the numerical simulation of the SASW measurements was used fir the calculation of the automatic calculation of the phase velocities and compared with the phase velocities by WinSASW employing the phase-unwrapping method. Also, the proposed method was applied to the real SASW measurements performed at$\bigcirc$$\bigcirc$area in GyeongGi-Do to see how the proposed method works with the real measurements.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2006.05a
• /
• pp.334-337
• /
• 2006

Generally, fragility curve has been used in predicting failure of structures due to seismic actions. In this research, the method of drawing fragility curve has been applied to evaluating success/failure of structures and satisfactory/unsatisfactory of concrete mixture performance based on material parameters. In the paper, a detailed explanation of the procedure of drawing fragility curve based on material parameter has been introduced. Fragility curve generating procedure includes generation of virtual data points from limited number of actual data points by bell curve implementation, determination of success/failure status of each data point by assigned criterion, and completion of final fragility curve. For practical applications, workability of concrete mixture content based on "unit water" has been used to obtain fragility curve. Detailed explanation of fragility curve drawing procedure for material parameters is presented.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2000.04a
• /
• pp.475-478
• /
• 2000

Recently encryption algorithms based on difficulties of factorization have been used with popularization. Prime number factorizations are progressed rapidly. In this paper, characteristics of elliptic curve are analyzed and generation of elliptic curves suitable for prime number factorization is discussed.

• 유체기계공업학회:학술대회논문집
• /
• 2001.11a
• /
• pp.143-148
• /
• 2001

In this study, to investigate the effect of the generalized performance curve on the performance prediction and to find the optimal ones, a systematic study is performed. For this purpose, we compared the influence of the stage performance curves with experimental data in multi-stage axial compressors. As a result, it is discovered that the optimal generalized performance curves vary according to the number of the stages in compressors. And we found that for a low-stage compressors, Muir's pressure coefficient curve gives the best prediction results at design rotational frequency regardless of the efficiency curve. On the other hand, for high-stage compressors, Stone's pressure coefficient curve gives the optimistic results about the performance prediction at design rotational frequency.

• Korean Journal of Computational Design and Engineering
• /
• v.4 no.1
• /
• pp.32-42
• /
• 1999

Approximate lofting or skinning is one of practical surface modeling techniques well used in CAD and reverse engineering applications. Presented in this paper is a method for approximately lofting a given set of curves wihin a specified tolereance. It is based on refitting input curves simultaneously on a common knot vector and interpolating them to get a resultant NURBS surface. A concept of reducing the number of interior knots of the common knot vector is well adopted to acquire more compact representation for the resultant surface. Energy minimization is newly introduced in curve refitting process to stabilize the solution of the fitting problem and get more fair curve. The proposed approximate lofting provides more smooth surface models and realizes more efficient data reduction expecially when the parameterization and compatibility of input curves are not good enough. The method has been successfully implemented in a new CAD/CAM product VX Vision? of Varimetrix Corporation.

• Proceedings of the KIEE Conference
• /
• 2003.11c
• /
• pp.445-448
• /
• 2003

This paper presents a novel iris recognition method using wavelet transform and curve simplification. One-dimensional signals, which are calculated over circles on the iris, are decomposed into a multiple frequency bands. Each decomposed signal is approximated by a piecewise linear curve connecting node points. The curve is simplified by progressively removing unimportant node points while keeping the shape of the curve. Finally, a small number of node points represent features of each signal. Experiment results show that the presented method results in good performance in various noise environments.

• Journal of the Korean Society of Safety
• /
• v.21 no.4 s.76
• /
• pp.42-48
• /
• 2006

The strategies of maintenance and operation are usually established based on the number of flaws and their size distribution obtained from nondestructive inspection in order to preserve safety of the plant. But non destructive inspection results are different from the physical flaws which really exist in the equipments. In case of a single inspection, it is easy to estimate the number of physical flaws using the POD curve. However, we may be faced with some difficulties in obtaining the number of physical flaws from the periodic in-service non destructive inspection data. In this study a simple method for estimating the number of physical flaws from periodic in-service nondestructive inspection data was proposed. In order to obtain the flaw growth history, the flaw growth was simulated using the Monte Carlo method and the flaw size and the corresponding POD value were obtained for each flaw at each periodic inspection time. The flaw growth rate used in the simulation was statistically calculated from the in-service inspection data. By repeating the simulation numerous flaw growth data could be generated and the effective POD curve was obtained as a function of flaw size. From the effective POD curve the number of physical flaws was obtained. The usefulness and convenience of the proposed method was evaluated from several applications and satisfactory results were obtained.

• Journal of Korea Water Resources Association
• /
• v.42 no.4
• /
• pp.297-307
• /
• 2009

For reliable water quality simulation by semi distributed model, accurate daily runoff simulation should have preceded. In this study, newly developed channel routing method which is nonlinear storage method is combination of Muskingum routing method and variable storage routing method and temporally weighted average curve number method were applied for effect analysis of water quality simulation. Developed modules, which are added in SWAT models and simulation, were conducted for the Chungju dam watershed. The simulation result by each module applied effect. As a result of analysis contribute water quality modeling, nonlinear storage method is more effective than temporally weighted average curve number method. Nutrient loading discharge was affected by development of runoff delaying from improvement of channel routing, because of characteristics of nonpoint source pollution.

• International journal of advanced smart convergence
• /
• v.10 no.3
• /
• pp.51-58
• /
• 2021

In this paper, we propose a deep learning classifier for estimating the number of layers in the Earth's structure. When installing a grounding system, knowledge of the subsurface in the area is absolutely necessary. The subsurface structure can be modeled by the earth parameters. Knowing the exact number of layers can significantly reduce the amount of computation to estimate these parameters. The classifier consists of a feedforward neural network. Apparent resistivity curves were used to train the deep learning classifier. The apparent resistivity at 20 equally spaced log points in each curve are used as the features for the input of the deep learning classifier. Apparent resistivity curve data sets are collected either by theoretical calculations or by Wenner's measurement method. Deep learning classifiers are coded by Keras, an open source neural network library written in Python. This model has been shown to converge with close to 100% accuracy.