• Journal of applied mathematics & informatics
• /
• v.23 no.1_2
• /
• pp.461-470
• /
• 2007

Parameter identification is studied in viscoelastic rods by solving an inverse problem numerically. The material properties of the rod, which appear in the constitutive relations, are recovered by optimizing an objective function constructed from reference strain data. The resulting inverse algorithm consists of an optimization algorithm coupled with a corresponding direct algorithm that computes the strain fields given a set of material properties. Numerical results are presented for two model inverse problems; (i)the effect of noise in the reference strain fields (ii) the effect of minimal reference data in space and/or time data.

• Journal of Applied Reliability
• /
• v.1 no.1
• /
• pp.17-33
• /
• 2001

The characteristics of Weibull distribution are investigated as a function of shape parameter. The statistical estimation methods of the shape parameter and statistical comparison methods of two or more shape parameters are studied. Assuming Weibull distribution, statistical analysis of bending-strength data of alumina titanium carbide ceramic matrix composites machined two different methods are performed.

• Communications for Statistical Applications and Methods
• /
• v.12 no.2
• /
• pp.413-423
• /
• 2005

Consider the problem of estimating the underlying regression function from a set of noisy data which is contaminated by a long tailed error distribution. There exist several robust smoothing techniques and these are turned out to be very useful to reduce the influence of outlying observations. However, no matter what kind of robust smoother we use, we should choose the smoothing parameter and relatively less attention has been made for the robust bandwidth selection method. In this paper, we adopt the idea of robust location parameter estimation technique and propose the robust cross validation score functions.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.1 no.1
• /
• pp.83-104
• /
• 1997

This paper studies parameter estimation for a first-order hyperbolic integro-differential equation modelling one-sex population dynamics. A second-order finite difference scheme is used to estimate parameters such as the age-specific death-rate and the age-specific fertility from fully discrete observations on the population. The function space parameter estimation convergence of this scheme is proved. Also, numerical simulations are performed.

• Journal of the Chungcheong Mathematical Society
• /
• v.33 no.4
• /
• pp.505-516
• /
• 2020

This study shows that image deblurring problems can be transformed into the multi-parameter Tikhonov type with multiple right hand sides. Also, this paper proposes the extension of the global generalized cross validation to obtain an appropriate choice of the regularization parameters for this problem. The experimental results of using the preconditioned Gl-CGLS algorithm were analyzed.

• Journal of the Korean Mathematical Society
• /
• v.52 no.3
• /
• pp.637-647
• /
• 2015

For generalized continued fraction (GCF) with parameter ${\epsilon}(k)$, we consider the size of the set whose partial quotients increase rapidly, namely the set $$E_{\epsilon}({\alpha}):=\{x{\in}(0,1]:k_{n+1}(x){\geq}k_n(x)^{\alpha}\;for\;all\;n{\geq}1\}$$, where ${\alpha}$ > 1. We in [6] have obtained the Hausdorff dimension of $E_{\epsilon}({\alpha})$ when ${\epsilon}(k)$ is constant or ${\epsilon}(k){\sim}k^{\beta}$ for any ${\beta}{\geq}1$. As its supplement, now we show that: $$dim_H\;E_{\epsilon}({\alpha})=\{\frac{1}{\alpha},\;when\;-k^{\delta}{\leq}{\epsilon}(k){\leq}k\;with\;0{\leq}{\delta}<1;\\\;\frac{1}{{\alpha}+1},\;when\;-k-{\rho}<{\epsilon}(k){\leq}-k\;with\;0<{\rho}<1;\\\;\frac{1}{{\alpha}+2},\;when\;{\epsilon}(k)=-k-1+\frac{1}{k}$$. So the bigger the parameter function ${\epsilon}(k_n)$ is, the larger the size of $E_{\epsilon}({\alpha})$ becomes.

• Journal of Mechanical Science and Technology
• /
• v.17 no.4
• /
• pp.484-491
• /
• 2003

For most of linear time-varying (LTV) systems, it is difficult to design time-varying controllers in analytic way. Accordingly, by approximating LTV systems as uncertain linear time-invariant, control design approaches such as robust control have been applied to the resulting uncertain LTI systems. In particular, a robust control method such as quantitative feedback theory (QFT) has an advantage of guaranteeing the frozen-time stability and the performance specification against plant parameter uncertainties. However, if these methods are applied to the approximated linear. time-invariant (LTI) plants with large uncertainty, the resulting control law becomes complicated and also may not become ineffective with faster dynamic behavior. In this paper, as a method to enhance the fast dynamic performance of LTV systems with bounded time-varying parameters, the approximated uncertainty of time-varying parameters are reduced by the proposed QFT parameter-scheduling control design based on radial basis function (RBF) networks.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.10
• /
• pp.1822-1828
• /
• 2008

This paper presents design parameter calculation methodology and its realization to construction for the 10/350${\mu}s$ lightning impulse current generator(ICG) modelled as double exponential function waveform with characteristic parameters ${\alpha},{\beta}$. Matlab internal function, "fzero" was applied to find ${\lambda}={\alpha}/{\beta}$ which is solution of nonlinear equation linearly related with two wave parameter $T_1$ and $T_2$. The calculation results for 10/350${\mu}s$ lightning impulse current show very good accuracy with error less 0.03%. Two type of 10/350${\mu}s$ ICGs based on the calculated design circuit parameters were fabricated by considering the load variation. One is applicable to the MOV based Surge protective device(SPD) for less 15 kA and the other is to test small resistive devices such as spark gap arrester and bonding device with maximum current capability 30 kA. The tested waveforms show error within 10% in comparison with the designed estimation and the waveform tolerance recommended in the IEC 61643-1 and IEC 60060-1.

• Journal of Korean Institute of Industrial Engineers
• /
• v.27 no.1
• /
• pp.47-53
• /
• 2001

In this paper, we compare the classification performances of both ensemble and clustering algorithms (Data Bagging, Variable Selection Bagging, Parameter Combining, Clustering) to logistic regression in consideration of various characteristics of input data. Four factors used to simulate the logistic model are (1) correlation among input variables (2) variance of observation (3) training data size and (4) input-output function. In view of the unknown relationship between input and output function, we use a Taguchi design to improve the practicality of our study results by letting it as a noise factor. Experimental study results indicate the following: When the level of the variance is medium, Bagging & Parameter Combining performs worse than Logistic Regression, Variable Selection Bagging and Clustering. However, classification performances of Logistic Regression, Variable Selection Bagging, Bagging and Clustering are not significantly different when the variance of input data is either small or large. When there is strong correlation in input variables, Variable Selection Bagging outperforms both Logistic Regression and Parameter combining. In general, Parameter Combining algorithm appears to be the worst at our disappointment.

• Journal of the Korean Society of Hazard Mitigation
• /
• v.10 no.6
• /
• pp.131-138
• /
• 2010

We verified the applicability of the developed universal model for the parameter estimation through the rainfall-runoff analysis at 16 watersheds. The existing parameter estimation equations derived from the restricted conditions sometimes, gave the meaningless results which cannot reflect the watershed characteristics and so have not widely used in the ungaged watershed. The values estimated from the developed universal model showed which are sensitive to variations of watershed characteristics. Wider applicability of SFM in ungaged watersheds is expected with the used of effective rainfall from CN method and the universal model.