• The Korean Journal of Applied Statistics
• /
• v.24 no.6
• /
• pp.1161-1168
• /
• 2011

The approximation for the distribution functions of nonparametric test statistics is a significant step in statistical inference. A rank sum test for dispersions proposed by Ansari and Bradley (1960), which is widely used to distinguish the variation between two populations, has been considered as one of the most popular nonparametric statistics. In this paper, the statistical tables for the distribution of the nonparametric Ansari-Bradley statistic is produced by use of polynomially adjusted normal approximation as a semi parametric density approximation technique. Polynomial adjustment can significantly improve approximation precision from normal approximation. The normal-polynomial density approximation for Ansari-Bradley statistic under finite sample sizes is utilized to provide the statistical table for various combination of its sample sizes. In order to find the optimal degree of polynomial adjustment of the proposed technique, the sum of squared probability mass function(PMF) difference between the exact distribution and its approximant is measured. It was observed that the approximation utilizing only two more moments of Ansari-Bradley statistic (in addition to the first two moments for normal approximation provide) more accurate approximations for various combinations of parameters. For instance, four degree polynomially adjusted normal approximant is about 117 times more accurate than normal approximation with respect to the sum of the squared PMF difference.

• The KSFM Journal of Fluid Machinery
• /
• v.3 no.1 s.6
• /
• pp.19-27
• /
• 2000

A numerical study is presented for analysis of three-dimensional incompressible turbulent flows in a multi-blade centrifugal fan. Reynolds-averaged Navier-Stokes equations with a standard $k-{\espilon}$ turbulence model are discretized with finite volume approximations. The computational area is divided into three blocks; inlet core, impeller and scroll parts, which are linked by a multi-block method. The flow inside of the fan is regarded as steady flow, and the mathematical models for the impeller forces were established from a cascade theory and measured data. Empirical coefficients are obtained comparing between computational and experimental results for the case without scroll, and are employed to simulate the flow through the impeller with scroll. In comparisons with experimental data, the validity of the mathematical models for the impeller forces was examined. The characteristics of the flow in the scroll were also discussed.

• The KSFM Journal of Fluid Machinery
• /
• v.7 no.3 s.24
• /
• pp.32-38
• /
• 2004

Shape of a multi-blades centrifugal fan is optimized by response surface method based on three-dimensional Navier-Stokes analysis. For numerical analysis, Reynolds-averaged Navier-Stokes equations with standard $k-{epsilon}$ turbulence model are transformed into non-orthogonal curvilinear coordinate system, and are discretized with finite volume approximations. Due to the large number of blades in this centrifugal fan, the flow inside of the fan is regarded as steady flow by introducing the impeller force models for economic calculations. Optimizations with and without constraints are carried out. Design variables, location of cur off, radius of cut off, expansion angle of scroll and width of impeller were selected to optimize the shapes of scroll and blades. Data points for response evaluations were selected by D-optimal design, and linear programming method was used for the optimization on the response surface. As a main result of the optimization, the efficiency was successfully improved. The correlation of efficiency with relative size of inactive zone at the exit of impeller is discussed as well as with average momentum fluxes in the scroll.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.47 no.11
• /
• pp.1-6
• /
• 2010

In Sensor Networks, the problem of finding the optimal routing path dynamically, which passes through all terminals or nodes once per each, may come up. Providing a generalized scheme of approximations that can be applied to the kind of problems, and formulating the bounds of the run time and the results of the algorithm made from the scheme, one may evaluate mathematically the routing path formed in a given network. This paper, dealing with Euclidean TSP(Euclidean Travelling Sales Person) that represents such problems, provides the scheme for constructing the approximated Euclidean TSP by parallel computing, and the ground for determining the difference between the approximated Euclidean TSP produced from the scheme and the optimal Euclidean TSP.

• The KIPS Transactions:PartB
• /
• v.9B no.3
• /
• pp.271-276
• /
• 2002

The test statistic in ANOVA tests has a single or doubly noncentral F distribution and the noncentral F distribution is applied to the calculation of the power functions of tests of general linear hypotheses. Although various approximations of noncentral F distribution are suggested, they are troublesome to compute. In this paper, the calculation of noncentral F distribution is applied to the neural network theory, to solve the computation problem. The neural network consists of the multi-layer perceptron structure and learning process has the algorithm of the backpropagation. Using fables and figs, comparisons are made between the results obtained by neural network theory and the Patnaik's values. Regarding of accuracy and calculation, the results by neural network are efficient than the Patnaik's values.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2004.10a
• /
• pp.496-500
• /
• 2004

A new, commercially available polishing process called magnetorheological finishing is used to polish and figure precision optics. To understand and model this process correctly it is important to determine the mechanical properties of the fluid under the influence of the magnetic field. Magnetorheological (MR) fluids are commonly modeled as Bingham fluids, so one of the essential properties to measure is the yield stress. Since MR fluids are inherently anisotropic, the yield stress will depend on the mutual orientation of the magnetic field and the direction of deformation. The relative orientation of the field and deformation in polishing does not coincide with common rheological setups, so a new rheometer has been designed and tested. This new magnetorheometer design has been shown to give correct stresses during calibration experiments using Newtonian fluids with a known viscosity. The measured stress has also been shown to have a magnitude consistent with published finite element approximations for magnetic fluids. The design of the instrument was complicated because of the requirements imposed upon the magnetic field, and the difficulty in satisfying the no slip boundary condition. Our results show the importance of having a homogeneous field in the test region during measurements. The solutions to these problems and discussion of the measurements on nonmagnetic and magnetic fluids are given.

• 유체기계공업학회:학술대회논문집
• /
• 1998.12a
• /
• pp.42-48
• /
• 1998

Numerical study is presented for the analysis of three-dimensional incompressible turbulent flows in multiblade centrifugal fan. Reynolds-averaged Navier-Stokes equations with standard k - $\epsilon$ turbulence model are transformed to non-orthogonal curvilinear coordinates, and are discretized with finite volume approximations. Linear Upwind Differencing Scheme(LUDS) is used to approximate the convection terms in the governing equations. SIMPLEC algorithm is used as a velocity-pressure correction procedure. The computational area is divided into three blocks; core, impeller and scroll, which are linked by multi-block method. The flow inside of the fan is regarded as steady flow, and mathematical formula established from the cascade theory and empirical coefficient are employed to simulate tile flow through the impeller. From comparisons between the computational results and the experimental data, the validity of the mathematical formula for the blade forces was examined and good results were obtained qualitatively. Hence, we can get the flow characteristics of multi-blade centrifugal fan and it will be a corner stone of the development of the multiblade centrifugal fan.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.28 no.9
• /
• pp.1314-1319
• /
• 2004

In this study, parameter optimization of micro-static mixer with a cantilever beam was accomplished for maximizing the mixing efficiency by using successive response surface approximations. Variables were chosen as the length of cantilever beam and the angle between horizontal and the cantilever beam. Sequential approximate optimization method was used to deal with both highly nonlinear and non-smooth characteristics of flow field in a micro-static mixer. Shape optimization problem of a micro-static mixer can be divided into a series of simple subproblems. Approximation to solve the subproblems was performed by response surface approximation, which does not require the sensitivity analysis. To verify the reliability of approximated objective function and the accuracy of it, ANOVA analysis and variables selection method were implemented, respectively. It was verified that successive response surface approximation worked very well and the mixing efficiency was improved very much comparing with the initial shape of a micro-static mixer.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.25 no.12
• /
• pp.67-73
• /
• 2011

A method obtaining position data(x, y) of object accurately is proposed by using a pair of ultrasonic sensors composed of one transmitter and two receivers. And the driving system which controls the light of flat LED lamp locally using array of ultrasonic sensors (3 transmitters and 6 receivers) is developed. As a result, measured values of y are relatively reliable due to its small average of absolute errors of 1.03[cm]. The measured values of x have average error of 8.52[cm], and it is a large value. However, the average error is decreased by 0.65[cm] after applying algorithm for error correction. The experiments to control the light of flat LED lamp locally with algorithm for error correction are carried out. From the result, measured values of x with average error of 0.97[cm] are obtained and they are very good approximations of actual values.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.7
• /
• pp.2371-2388
• /
• 2015

In this paper, we investigate the performance evaluation of three dimensional (3D) multiple-input multiple-output (MIMO) systems with an adjustable base station (BS) antenna tilt angle and zero-forcing (ZF) receivers in Ricean/Lognormal fading channels. In particular, we take the lognormal shadow fading, 3D antenna gain with antenna tilt angle and path-loss into account. First, we derive a closed-form lower bound on the sum rate, then we obtain the optimal BS antenna tilt angle based on the derived lower bound, and finally we present linear approximations for the sum rate in high and low-SNR regimes, respectively. Based on our analytical results, we gain valuable insights into the impact of key system parameters, such as the BS antenna tilt angle, the Ricean K-factor and the radius of cell, on the sum rate performance of 3D MIMO with ZF receivers.