• 대한수학회논문집
• /
• 제25권1호
• /
• pp.11-18
• /
• 2010

The work in this paper was motivated by [3], where Eschenbach and Li listed four 4 by 4 sign patterns, conjectured to be nilpotent sign patterns of nilpotence index at least 3. These sign patterns with no zero entries, called full sign patterns, are shown to be potentially nilpotent of nilpotence index 3. We also generalize these sign patterns of order 4 so that we provide classes of n by n sign patterns of nilpotence indices at least 3, if they are potentially nilpotent. Furthermore it is shown that if a full sign pattern A of order n has nilpotence index k with $2{\leq}k{\leq}n-1$, then sign pattern A has nilpotent realizations of nilpotence indices k, k + 1, $\ldots$, n. Hence, the four 4 by 4 sign patterns in [3, page 91] also allow nilpotent realizations of nilpotence index 4.

• Structural Engineering and Mechanics
• /
• 제8권5호
• /
• pp.453-464
• /
• 1999

Safety monitoring systems of structures generally resort to detecting possible changes of dynamic system parameters. Sensitivity analysis of these dynamic system parameters may implement these techniques. Conventional structural eigenvalue problems are discussed in the scope of those systems with deterministic parameters. Large and flexible structures, such as suspension bridges, actually possess stochastic material properties and these random properties unavoidably affect the dynamic system parameters. The sensitivity matrix of structural modal parameters to basic design variables has been established in this paper. Moreover, second order statistics of natural frequencies due to the randomness of material properties have been discussed. It is concluded from numerical analysis of a modem suspension bridge that although the second order statistics of frequencies are small relatively to the change of basic design variables, such as density of mass and modulus of elasticity, the sensitivities of modal parameters to these variables at different locations change in magnitude.

• Journal of the Korean Data and Information Science Society
• /
• 제18권2호
• /
• pp.279-287
• /
• 2007

Data fusion is the process of combining multiple data in order to produce information of tactical value to the user. Data fusion is generally defined as the use of techniques that combine data from multiple sources and gather that information in order to achieve inferences. Data fusion is also called data combination or data matching. Data fusion is divided in five branch types which are exact matching, judgemental matching, probability matching, statistical matching, and data linking. In this paper, we develop was macro program for statistical matching which is one of five branch types for data fusion. And then we apply data fusion and association rule techniques to environmental data.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2002년도 학술대회지
• /
• pp.827-830
• /
• 2002

A direct numerical simulation is performed for turbulent heat transfer in a concentric annulus at $Re_{Dh}=8900\;and\;Pr=0.71$ for two radius ratios ($R_{1}/R_{2}=0.1\;and\;0.5$) and $q^{\ast}=1.0$. Main emphasis is placed on the transverse curvature effect on near-wall turbulent thermal structures. Near-wall turbulent structures close to the inner and outer walls are scrutinized by computing the lower-order statistics. The fluctuating temperature variance and turbulent heat flux budgets are illustrated to confirm the results of the lower-order statistics. The present numerical results show that the turbulent structures near the outer wall are more activated than those near the inner wall, which may be attributed to the different vortex regeneration processes between the inner and outer walls.

• 한국신뢰성학회:학술대회논문집
• /
• 한국신뢰성학회 2002년도 정기학술대회
• /
• pp.297-302
• /
• 2002

An efficient method is developed to obtain the maximum capacity flow for a network when its simple paths are known. Most of the existing techniques need to convert simple paths into minimal cuts, or to determine the order of simple paths to be applied in the process to reach the correct result. In this paper, we propose a method based on the concepts of signed simple path and signed flow defined in the text. Our method involves a fewer number of arithmetic operations at each iteration, and requires fewer iterations in the whole process than the existing methods. Our method can be easily extended to a mixed network with a slight modification. Furthermore, the correctness of our method does not depend on the order of simple paths to be applied in the process.

• Journal of the Korean Statistical Society
• /
• 제32권4호
• /
• pp.337-357
• /
• 2003

This paper introduces new measures for the stability of slope estimation on the second order response surface at a point and on a sphere. As a measure of point stability of slope estimation, we suggest a point dispersion measure of slope variances over all directions at a point. A spherical dispersion measure is also proposed as a measure of spherical stability of slope estimation on each sphere. Some designs are studied to explore the usefulness of the proposed measures. Using the point dispersion measure, another concept of slope rotatability called equally-stable slope rotatability is proposed as a useful property of response surface designs. We provide a set of conditions for a design to have equally-stable slope rotatability.

• Journal of the Korean Statistical Society
• /
• 제25권3호
• /
• pp.347-358
• /
• 1996

A new test for goodness-of-fit is presented. It is a modification of a test of LaRiccia (1991). These tests are applicable to continuous lo-cation/scale models. The new test statistic is based on a few selected order statistics taken from the sample, while the LaRiccia test is based directly on the full sample. Each test embeds the hypothesized model in a larger linear model and proceeds to test the goodness-of-fit hy-pothesis by testing the coefficients of this linear model appropriately. The general theory is presented. The tests are compared via computer simulation to a related test of Ali and Umbach (1989) for distributions that could be used as lifetime models. An important aspect of all these tests is that only standard $X_2$ tables are used. Selection of the spacings of the order statistics is discussed.

• Communications for Statistical Applications and Methods
• /
• 제23권2호
• /
• pp.147-161
• /
• 2016

This paper introduces the four parameter new generalized inverse Weibull distribution and investigates the potential usefulness of this model with application to reliability data from engineering studies. The new extended model has upside-down hazard rate function and provides an alternative to existing lifetime distributions. Various structural properties of the new distribution are derived that include explicit expressions for the moments, moment generating function, quantile function and the moments of order statistics. The estimation of model parameters are performed by the method of maximum likelihood and evaluate the performance of maximum likelihood estimation using simulation.

• 품질경영학회지
• /
• 제27권2호
• /
• pp.237-250
• /
• 1999

Response surface analysis has been a popular tool conducted by engineers in many processes. In this paper, response surface function, named partial least squares response surface function is proposed. Partial least squares response surface function is a function of partial least squares components and the response surface modeling is used in either a first-order or a second-order model. Also, this approach will have the engineers be able to do the response surface modeling and the process optimization even when the number of experimental runs is less than the number of model parameters. This idea is applied to the nondesign data and an application of partial least squares response surface function to the process optimization is considered.

• 대한전자공학회논문지
• /
• 제27권1호
• /
• pp.112-119
• /
• 1990

Two adaptation procedures are proposed for image restoration, in which the window shapes of OS filters are changed according to the order statistics of signal. In the first procedure, the 2-dimensional window is the union of the 1-dimensional subwindows whose sizes are fixed. In the second procedure, the 2-dimensional window is the union of the 1-dimensional subwindows whose sizes are variable. Compared with existing procedures, our adaptation procedures using order statistics are computationally efficient. Simulation results show that the filters with adaptive window shapes have good performance for the preservation of edges and details of image, and the noise suppression.