• Journal of the Korean Data and Information Science Society
• /
• 제2권
• /
• pp.45-62
• /
• 1991

신뢰수명의 검정이나 임상실험에서 대상물에 대한 관측치를 충분히 얻기 위해서는 많은 시간과 경비가 필요하거나 현실적인 제한으로 인하여 관측이 불가능한 경우가 흔히 있다. 이러한 이유로 점진적 임의중단법에 의하여 얻어진 관측치를 이용한 생존함수의 추정은 현실적으로 매우 중요하다고 하겠다. 연구에서는 점진적 임의중단된 자료를 기초로 스플라인함수를 이용하여 생존함수의 비모수적 최우추정량을 제안하고 그것의 성질과 효율성을 비교 연구한다.

• Journal of the Korean Statistical Society
• /
• 제24권1호
• /
• pp.233-242
• /
• 1995

In this paper we will prove the existence and uniqueness of the smoothest density with prescribed moments. The space of functions considered is the Sobolev space $W^2_m[0,1]$ and the target functional to be minimized is the seminorm $$\mid$$\mid$f^{(m)}$\mid$$\mid$_{L^2}$, which measures the roughness of the function f.

• Journal of Advanced Marine Engineering and Technology
• /
• 제32권5호
• /
• pp.753-758
• /
• 2008

A CAM program for forming design Automated of CAM for Spring using car was developed in this study. This program was written in AUTO LISP on the AUTO CAD system with a personal. An approach to the system is based on the kinemateic of the object function. We make a determination of an cam programming. A CAM spline is continuous in displacement. velocity and acceleration. This program which can design Coil Spring shapes will successfully support CAN designing and manufacturing for Small & Medium companies.

• 한국주조공학회지
• /
• 제5권2호
• /
• pp.125-136
• /
• 1985

The most desirable method for the solution of solidification problems in castings must be the one which combines accuracy, simplicity, and low computer usage cost. The purpose of the present study is to develop a method which includes all these advantages. The purpose of the present research was approached by the introduction of two methods: (1) A pour-out test, employed with very high purity aluminum, for the purpose of obtaining accurate solidification data in L-sections; and (2) an numerical technique, using the cubic spline function for defining solidification curves.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2008년도 춘계종합학술대회 A
• /
• pp.205-208
• /
• 2008

수치해석에서는 온도, 압력, 비체적, 엔탈피, 엔트로피 등의 수치값이 필요하다. 그런데 증기표의 대부분의 열역학적 성질들은 측정된 값이기 때문에 기본적으로 측정 오차를 가지고 있다. 본 연구에서는 압력 기준의 물의 포화 상태에 대해, 난수를 발생시켜 적절한 크기로 조절한 다음 원래의 성질들에 더하여 인위적으로 노이즈가 포함된 데이터를 만들었다. 이 데이터를 신경회로망과 스플라인 보간법으로 함수 근사를 하였다. 해석 결과 신경회로망이 2차 스플라인 보간법보다 훨씬 더 적은 백분율 오차를 보였으며 이로부터 신경회로망이 측정 오차의 영향을 적게 받는 함수 근사에 적절한 방법임을 확인하였다.

• Geomechanics and Engineering
• /
• 제2권2호
• /
• pp.141-156
• /
• 2010

In the framework of meshfree methods, a new methodology is developed based on radial point interpolation method (RPIM). This methodology is applied to a one-dimensional contaminant transport modelling in the saturated porous media. The one-dimensional form of advection-dispersion equation involving reactive contaminant is considered in the analysis. The Galerkin weak form of the governing equation is formulated using 1D meshfree shape functions constructed using thin plate spline radial basis functions. MATLAB code is developed to obtain the numerical solution. Numerical examples representing various phenomena, which occur during migration of contaminants, are presented to illustrate the applicability of the proposed method and the results are compared with those obtained from the analytical and finite element solutions. The proposed RPIM has generated results with no oscillations and they are insensitive to Peclet constraints. In order to test the practical applicability and performance of the RPIM, three case studies of contaminant transport through the landfill liners are presented. A good agreement is obtained between the results of the RPIM and the field investigation data.

• 한국멀티미디어학회논문지
• /
• 제20권8호
• /
• pp.1406-1420
• /
• 2017

Artificial neural networks (ANNs) play an important role in the fields of function approximation, prediction, and classification. ANN performance is critically dependent on the input parameters, including the number of neurons in each layer, and the optimal values of weights and biases assigned to each neuron. In this study, we apply the particle swarm optimization method, a popular optimization algorithm for determining the optimal values of weights and biases for every neuron in different layers of the ANN. Several regression models, including general linear regression, Fourier regression, smoothing spline, and polynomial regression, are conducted to evaluate the proposed method's prediction power compared to multiple linear regression (MLR) methods. In addition, residual analysis is conducted to evaluate the optimized ANN accuracy for both training and test datasets. The experimental results demonstrate that the proposed method can effectively determine optimal values for neuron weights and biases, and high accuracy results are obtained for prediction applications. Evaluations of the proposed method reveal that it can be used for prediction and estimation purposes, with a high accuracy ratio, and the designed model provides a reliable technique for optimization. The simulation results show that the optimized ANN exhibits superior performance to MLR for prediction purposes.

• 한국기계가공학회지
• /
• 제9권6호
• /
• pp.87-94
• /
• 2010

In this paper, a new simulator of the biped robot for planning walking patterns was showed. And this simulator(MHBiped) is able to not only visualize the plan of patterns but also verify whether a biped robot design is suitable. In addition, MHBipd can modify various kinds of walking parameters and the trajectory of biped robot. Therefore, a new biped robot can be designed easily by the this simulator before you apply to a robot. As a result, a well-balanced parameters of walking patterns watching the movement of CG and ZMP can be obtained. Walking patterns should be changed according to both the existence of obstacles and conditions of ground and it can be described by the trajectory of hip and ankles. All those trajectorys can be also obtained by the cubic spline functions and the way of modeling walking patterns. The results of simulator, the movement function of CG and ZMP, the cubic spline functions and modeling of biped robot were introduced in this paper. And the effectiveness of this simulator was confirmed by the simulations.

• 한국정밀공학회지
• /
• 제29권4호
• /
• pp.441-448
• /
• 2012

This paper describes the optimal home positioning algorithm of Eclipse-II, a new conceptual parallel mechanism for motion simulator. Eclipse-II is capable of translation and 360 degrees continuous rotation in all directions. In unexpected situations such as emergency stop, riders have to be resituated as soon as possible through a shortest translational and rotational path because the return paths are not unique in view of inverse kinematic solution. Eclipse-II is man riding. Therefore, the home positioning is directly related to the safety of riders. To ensure a least elapsed time, ZYX Euler angle inverse kinematics is applied to find an optimal home orientation. In addition, the subsequent decrease of maximum acceleration and jerk values is achieved by combining the optimal return path function with cubic spline, which consequently reduces delivery force and vibration to riders.

• 한국멀티미디어학회논문지
• /
• 제13권2호
• /
• pp.196-204
• /
• 2010

본 논문에서는 핸드폰 조립 전의 키패드의 개별버튼에 불량 검사방법을 제안한다. 제안한 알고리즘은 키패드의 영역 분할, 폰트 영역 분할, 이동 및 회전의 처리과정을 통해, 동일 색상의 검사 및 등급 분류, 폰트 검사, 스크래치 검사로 이루어져 있다. 특히, 본 논문에서 제안한 영역분할 방법은 기존의 단순 문턱치를 기반으로 키패드의 경사나 곡면 모양에 기인한 휘도 변화에 대응하기 위해, 패드 영역만을 B-spline으로 근사화하여, 각 화소마다 다른 표면 문턱치를 적용하는 방법을 제시한다. 또한, 키패드의 회전 정보를 고유치 및 고유벡터를 사용하여 매우 빠르고 효율적으로 구하는 방법을 제시한다. 실험결과 제안한 방법은 실제 인라인 공정상에 적용하여 실험결과 우수한 성능을 보여준다.