• Communications for Statistical Applications and Methods
• /
• 제18권3호
• /
• pp.267-275
• /
• 2011

According to recent studies, Bayesian information criteria(BIC) is proposed to determine the structural dimension of the central subspace through sliced inverse regression(SIR) with high-dimensional predictors. The BIC may be useful in K-means clustering inverse regression(KIR) with high-dimensional predictors. However, the direct application of the BIC to KIR may be problematic, because the slicing scheme in SIR is not the same as that of KIR. In this paper, we present empirical penalty term studies of BIC in KIR to identify the most appropriate one. Numerical studies and real data analysis are presented.

• 한국인구학
• /
• 제12권2호
• /
• pp.56-68
• /
• 1989

The methodology in historical demography comprises the three core areas the family reconstitution method at the Institut National d' Ftudes Demographiques(I.N.E.D), the back projection at the Cambridge Group for the History of PopuJation and Social Struc-ture(HPSS). and the household-pattern analysis at the Cambridge Group and at the California Institute of Technology. The paper presents an outline of the family reconstitu-tion method and discusses the problems, both theoretical and methodological, arising from the problematic back projection vis-a-vis the usual inverse projection developed by Ronald D. Lee at Berkeley. Recent developments in the tield of the generalized inverse projection method designed 10 supplement the defects in the back projection and the inverse projection are presented. and for ease of explanation of the parish register data for the family reconstitution form (FRE). pre-modern Korean household register data are presented along with the parish register data of England and Wales that constitute the backbone of historical demography in pre-modern Europe. Possibilities of exploring the household pattern analysis method based on the Laslett-Hammel classification scheme for the mid-eighteenth-century Korean household register data are suggested.

• 대한기계학회논문집A
• /
• 제22권3호
• /
• pp.483-492
• /
• 1998

An inverse analysis been applied to obtain the flow stress of the material. In this method, a ring-shaped specimen is compressed between two flat tools. This procedure employs, as the object function of inverse analysis, the balance of measured loads and reaction forces calculated by using rigid-plastic finite element method. The balance is explicit scalar function of flow stress which is a function of some unknown constants. For minimizing the balance, Newton-Raphon scheme is used. The friction factor, m, between flat tools and the specimen is determined by using friction area-divided method. The proposed method allows an accurate identification by avoiding the usual assumptions made in order to convert experimental measures into stress-strain relation. In this paper, the proposed method is numerically tested. A commercial pure aluminum was selected, as an example, to apply the method and the results are compared with stress-strain relation obtained by experiments.

• 한국기계가공학회지
• /
• 제13권5호
• /
• pp.90-97
• /
• 2014

This paper presents a simplified dynamics model, dynamics simulations, and computed torque control experiments of the Delta high-speed parallel robot. Using the typical Newton-Euler method, a simplified but accurate dynamics model with practical assumptions is derived. Accuracy and fast calculations of the dynamics are essential in the computed torque control for high-speed applications. It was found that the simplified dynamics equation is in very god agreement with the ADAMS model, and the calculation time of the inverse kinematics and inverse dynamics is about 0.04 msec. From the dynamics simulations, the cycle trajectory along the y-axis requires less peak motor torque and a lower angular velocity and less power than that along the x-axis. The computed torque control scheme can reduce the position error by half as compared to a PD control scheme. Finally, the developed Delta parallel robot prototype, half the size of the ABB Flexpicker robot, can achieve a cycle time of 0.43 sec with a 1.0kg payload.

• Journal of Power Electronics
• /
• 제17권3호
• /
• pp.744-755
• /
• 2017

With the high penetration of various sustainable energy sources, the control and protection of Microgrids has become a challenging problem considering the inherent current limitation feature of inverter-based Distributed Generators (DGs) and the bidirectional power flow in Microgrids. In this paper, a hybrid control and protection scheme is proposed, which combines the traditional inverse-time overcurrent protection with the biased differential protection for different feeders with different kinds of loads. It naturally accommodates various control strategies such as P-Q control and V-f control. The parameter settings of the protection scheme are analyzed and calculated through a fast Fourier transform algorithm, and the stability of the control strategy is discussed by building a small signal model in MATLAB. Different operation modes such as the grid-connected mode, the islanding mode, and the transitions between these two modes are ensured. A Microgrid model is established in PSCAD and the analysis results show that a Microgrid system can be effectively protected against different faults such as the single phase to ground and the three phase faults in both the grid-connected and islanded operation modes.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1993년도 한국자동제어학술회의논문집(국내학술편); Seoul National University, Seoul; 20-22 Oct. 1993
• /
• pp.760-765
• /
• 1993

A multivariable learning control is designed in frequency domain. A general to of feedback assisted learning scheme is considered and an inverse model based learning algorithm is derived through convergence analysis in frequency domain. Performance of the proposed control method is evaluated through numerical simulation.

• 응용통계연구
• /
• 제34권6호
• /
• pp.957-968
• /
• 2021

역중도절단확률가중(inverse censoring probability weighting, ICPW)은 생존분석에서 흔히 사용되는 방법이다. 중도절단 회귀모형과 같은 ICPW 방법의 응용에 있어서 중도절단 확률의 정확한 추정은 핵심적인 요소라고 할 수 있다. 본 논문에서는 중도절단 확률의 추정이 ICPW 기반 중도절단 회귀모형의 성능에 어떠한 영향을 주는지 모의실험을 통하여 알아보았다. 모의실험에서는 Kaplan-Meier 추정량, Cox 비례위험(proportional hazard) 모형 추정량, 그리고 국소 Kaplan-Meier 추정량 세 가지를 비교하였다. 국소 KM 추정량에 대해서는 차원의 저주를 피하기 위해 공변량의 차원축소 방법을 추가적으로 적용하였다. 차원축소 방법으로는 흔히 사용되는 주성분분석(principal component analysis, PCA)과 절단역회귀(sliced inverse regression)방법을 고려하였다. 그 결과 Cox 비례위험 추정량이 평균 및 중위수 중도절단 회귀모형 모두에서 중도절단 확률을 추정하는 데 가장 좋은 성능을 보여주었다.

• International Journal of Automotive Technology
• /
• 제2권3호
• /
• pp.117-122
• /
• 2001

For a dynamic analysis of a constrained multibody system, it is necessary to have a routine for satisfying kinematic constraints. LU decomposition scheme, which is used to divide coordinates into dependent and independent coordinates, is efficient but has great difficulty near the singular configuration. Other method such as the projection matrix, which is more stable near a singular configuration, takes longer simulation time due to the large amount of calculation for decomposition. In this paper, the row space and the null space of the Jacobian matrix are proposed by using the pseudo-inverse method and the projection matrix. The equations of the motion of a system are replaced with independent acceleration components using the null space of the Jacobian matrix. Also a new hybrid method is proposed, combining the LU decomposition and the projection matrix. The proposed hybrid method has following advantages. (1) The simulation efficiency is preserved by the LU method during the simulation. (2) The accuracy of the solution is also achieved by the projection method near the singular configuration.

• 산업기술연구
• /
• 제26권B호
• /
• pp.191-198
• /
• 2006

In this paper, we focus on a question. Which is better between time-based location update method and movement-based location update method? Or, does any other method combining the two methods show better performance? For the question, we propose a hybrid location update scheme, which integrates the time-based and the movement-based methods. In the proposed scheme, a mobile terminal updates its location after n cell boundary crossing and a time interval of T, or the inverse. We derive an analytical solution for the performance of the hybrid scheme with exponential cell resident time. From the numerical analysis, we conclude that the movement-based method seems to have better performance than the time-based and hybrid methods, that is the optimal costs occur at T=0.

• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 2005년도 춘계학술대회 논문집
• /
• pp.138-143
• /
• 2005

Finite Element analysis is widely applied to elevated temperature forging processes and shows a lot of information of plastic deformation such as strain, stress, defects, damages and temperature distributions. In highly elevated temperature deformation processes, temperature of material and tool have significant influence on tool life, deformation conditions and productivities. To predict temperature related properties accurately, adequate coefficients of not only contact heat transfer between material and dies but also convection heat transfer due to coolants are required. In most F.E analysis, too higher value of contact heat transfer coefficient is usually applied to get acceptable temperature distribution of tool. For contact heat transfer coefficients between die and workpiece, accurate values were evaluated with different pressure and lubricants conditions. But convection heat transfer coefficients have not been investigated for forging lubricants. In this research, convection heat transfer coefficients for cooling by emulsion lubricants are suggested by experiment and Inverse method. To verify acquired convection and contact heat transfer coefficients, tool temperature was measured for the comparison between measured tool temperature and analysis results. To increase analysis accuracy, repeated analysis scheme was applied till temperature of the tool got to be in the steady-state conditions. Verification of heat transfer coefficients both contact and convection heat transfer coefficients was proven with good accordance between measurement and analysis.