• Proceedings of the KSME Conference
• /
• 2008.11b
• /
• pp.2401-2405
• /
• 2008

In the present study, a three-dimensional least square/level set based two-phase flow code was developed for the simulation of three-dimensional sloshing problems using finite element discretization. The present method can be utilized for the analysis of a free surface flow problem in a complex geometry due to the feature of FEM. Since the finite element method is employed for the spatial discretization of governing equations, an unstructured mesh can be naturally adopted for the level set simulation of a free surface flow without an additional load for the code development except that solution methods of the hyperbolic type redistancing and advection equations of the level set function should be devised in order to give a bounded solution on the unstructured mesh. From the numerical experiments of the present study, it is shown that the proposed method is both robust and accurate for the simulation of three-dimensional sloshing problems.

• 한국전산유체공학회:학술대회논문집
• /
• 2007.04a
• /
• pp.17-22
• /
• 2007

Chimera grid Method is widely used in Computational Fluid Dynamics due to its simplicity in constructing grid system over complex bodies. Especially, Chimera grid method is suitable for unsteady flow computations with bodies in relative motions. However, interpolation procedure for ensuring continuity of solution over overlapped region fails when so-call orphan cells are present. We have adopted MLS(Moving Least Squares) method to replace commonly used linear interpolations in order to alleviate the difficulty associated with orphan cells. MSL is one of interpolation methods used in mesh-less methods. A number of examples with MLS are presented to show the validity and the accuracy of the method.

• Communications for Statistical Applications and Methods
• /
• v.4 no.2
• /
• pp.541-548
• /
• 1997

In regression model with nested error structure interval estimations on regression coefficients in different stages are proposed. Ordinary least square estimators and generalized least square estimators of the regression coefficients in this model are derived for between and within group model. The confidence intervals are dervied by using independent idstributional properties between regression coefficient estimators and quadratic froms obtained from the model.

• Journal of electromagnetic engineering and science
• /
• v.4 no.2
• /
• pp.93-96
• /
• 2004

In this paper we introduce a new escalator(ESC) structure-based adaptation algorithm. The proposed algorithm is independent of eigenvalues spread ratio(ESR) of channel and has faster convergence speed than that of the conventional ESC algorithms. This algorithm combines the fast adaptation ability of least square methods and the orthogonalization property of the ESC structure. From the simulation results the proposed algorithm shows superior convergence speed and no slowing down of convergence speed when we increase the ESR of the channel.

• 한국정보통신설비학회:학술대회논문집
• /
• 2008.08a
• /
• pp.190-192
• /
• 2008

This paper shows the formulation of fast moving least square reproducing kernel method (FMLSRKM) which is a kind of meshfree methods. FMLSRKM has some advantages compared to conventional numerical techniques such as finite element method. For simple analysis on a rectangular waveguide, point collocation scheme is introduced and applied.

• Journal of Sensor Science and Technology
• /
• v.26 no.4
• /
• pp.259-265
• /
• 2017

Magnetometer calibration must be performed before the use of three-axis magnetometers to ensure the accuracy of orientation estimation. Recently, one of the most popular calibration approaches is the ellipsoid fitting technique due to its high performance in calibration. To date, in fact, performances of the existing ellipsoid fitting methods have been evaluated with full range rotation data. However, in case of the calibration of magnetometers attached to vehicles, ships, and planes, it is very difficult to collect the full range rotation data since their allowable ranges in terms of roll and pitch are limited to small. This constraint may result in serious performance degradation of some ellipsoid fitting algorithms. Therefore, to be practical, this paper proposes a weighted least square-based magnetometer calibration method that is robust in roll-pitch limited conditions. Furthermore, the proposed method is a linear approach and thus is free from the well-known initial value issue in nonlinear approaches. Experimental results show the superiority of the proposed method to other ellipsoid-fitting calibration methods.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.5
• /
• pp.1151-1160
• /
• 2014

In DNA microarray studies, the number of genes far exceeds the number of samples and the gene expression measures are highly correlated. Partial least squares regression (PLSR) is one of the popular methods for dimensional reduction and known to be useful for the classifications of microarray data by several studies. In this study, we suggest a modified version of the partial least squares regression to analyze gene expression data with survival information. The method is designed as a new gene selection method using PLSR with an iterative procedure of imputing censored survival time. Mean square error of prediction criterion is used to determine the dimension of the model. To visualize the data, plot for variables superimposed with samples are used. The method is applied to two microarray data sets, both containing survival time. The results show that the proposed method works well for interpreting gene expression microarray data.

• Structural Engineering and Mechanics
• /
• v.25 no.4
• /
• pp.467-480
• /
• 2007

Two methods were developed for analyzing problems with two adjacent sub-domains modeled by different kinds of elements in finite element method. Each sub-domain can be defined independently without the consideration of equivalent division with common nodes used for the interface. These two methods employ an individual interface to accomplish the compatibility. The MLSA method uses the moving least square approximation which is the basic formulation for Element Free Galerkin Method to formulate the interface. The displacement field assumed by this method does not pass through nodes on the common boundary. Therefore, nodes can be chosen freely for this method. The results show that the MLSA method has better approximation than traditional methods.

• The Korean Journal of Applied Statistics
• /
• v.29 no.4
• /
• pp.595-611
• /
• 2016

We use nonlinear regression models (such as the Hill Model) when we analyze data in toxicology and/or pharmacology. In nonlinear regression models an estimator of parameters and estimation of measurement about uncertainty of the estimator are influenced by the variance structure of the error. Thus, estimation methods should be different depending on whether the data are homoscedastic or heteroscedastic. However, we do not know the variance structure of the error until we actually analyze the data. Therefore, developing estimation methods robust to the variance structure of the error is an important problem. In this paper we propose a method to estimate parameters in nonlinear regression models based on a preliminary test. We define an estimator which uses either the ordinary least square estimation method or the iterative weighted least square estimation method according to the results of a simple preliminary test for the equality of the error variance. The performance of the proposed estimator is compared to those of existing estimators by simulation studies. We also compare estimation methods using real data obtained from the National Toxicology program of the United States.

• 한국전산유체공학회:학술대회논문집
• /
• 2009.04a
• /
• pp.239-244
• /
• 2009

The gridless (or meshfree) methods, such as MPS, SPH, FPM an so forth, are feasible and robust for the problems with moving boundary and/or complicated boundary shapes, because these methods do not need to generate a grid system. In this study, a gridless solver, which is based on the combination of moving least square interpolations on a cloud of points with point collocation for evaluating the derivatives of governing equations, is presented for two-dimensional unsteady incompressible Navier-Stokes problem in the low Reynolds number. A MAC-type algorithm was adopted and the Poission equation for the pressure was solved by successively in the moving least square sense. Some weighing functions were tested in order to investigate the up-winding effect for the convection term. Some typical problems were solved by the presented solver for the validation and the results obtained were compared with analytic solutions and the numerical results by conventional CFD methods, such as FVM.