• The Korean Journal of Applied Statistics
• /
• v.6 no.2
• /
• pp.253-263
• /
• 1993

Recent interest in Taguchi's methods have led to developments of joint modelling of the mean and dispersion in generalized linear models. Since a single data transformation cannot produce all the necessary conditions for an analysis, for the analysis of the Taguchi data, the use of the generalized linear models is preferred to a commonly used data transformation method. In this paper, we will illustrate this point and provide GLIM macros to implement the joint modelling of the mean and dispersion in generalized linear models.

• Journal of Korean Society for Quality Management
• /
• v.21 no.1
• /
• pp.161-170
• /
• 1993

This study attempts to present the linear regression analysis that involves more than one regressor variable, because regression analysis is the most widely used statistical technique for describing, predicting and estimating the relationships between given data. The model of multiple linear regression may be solved directly by the two linear programming methods, i.e., to minimize the sum of the absolute deviation (MSD) and to minimize the maximum deviation(MMD). In addition, some results was compared to each techniques for accuracy and tested to the validity of statistical meaning.

• Archives of Pharmacal Research
• /
• v.11 no.2
• /
• pp.99-113
• /
• 1988

Quantitation of ethenzamide, isopropylantipyrine and caffeine takes about 41 hrs by conventional GC method. Quantitation of allylisoprorylacetylurea takes about 40 hrs by conventional UV method. But quantitation of them takes about 6 hrs by DRIFT developing method. Each standard and sample sieved, powdered and acquired DRIFT spectrum. Out of them peak of each component was selected and ratio of each peak to standard peak was acquired, and then linear stepwise multiple regression was performed with these data and concentration. Reflectance value, Kubelka-Munk equation and Inverse-Kubelka-Munk equation were modified by us. Inverse-Kubelka-Munk equation completed the deficit of Kubelka-Munk equation. Correlation coefficients acquired by conventioanl GC and UV against DRIFT were more than 0.95.

• Communications for Statistical Applications and Methods
• /
• v.31 no.4
• /
• pp.393-408
• /
• 2024

The sparse linear discriminant analysis can be incorporated into the penalized linear regression framework, but most studies have been limited to specific convex penalties, including the least absolute selection and shrinkage operator and its variants. Within this framework, concave penalties can serve as natural counterparts of the convex penalties. Implementing the concave penalized direction vector of discrimination appears to be straightforward, but developing its theoretical properties remains challenging. In this paper, we explore a class of concave penalties that covers the smoothly clipped absolute deviation and minimax concave penalties as examples. We prove that employing concave penalties guarantees an oracle property uniformly within this penalty class, even for high-dimensional samples. Here, the oracle property implies that an ideal direction vector of discrimination can be exactly recovered through concave penalized least squares estimation. Numerical studies confirm that the theoretical results hold with finite samples.

• Structural Engineering and Mechanics
• /
• v.56 no.1
• /
• pp.27-47
• /
• 2015

In this study, the accuracy and reliability of fully nonlinear method against equivalent linear method for dynamic analysis of soil-structure interaction is investigated comparing the predicted results of both numerical procedures with the results of experimental shaking table tests. An enhanced numerical soil-structure model has been developed which treats the behaviour of the soil and the structure with equal rigour. The soil-structural model comprises a 15 storey structural model resting on a soft soil inside a laminar soil container. The structural model was analysed under three different conditions: (i) fixed base model performing conventional time history dynamic analysis, (ii) flexible base model (considering full soil-structure interaction) conducting equivalent linear dynamic analysis, and (iii) flexible base model performing fully nonlinear dynamic analysis. The results of the above mentioned three cases in terms of lateral storey deflections and inter-storey drifts are determined and compared with the experimental results of shaking table tests. Comparing the experimental results with the numerical analysis predictions, it is noted that equivalent linear method of dynamic analysis underestimates the inelastic seismic response of mid-rise moment resisting building frames resting on soft soils in comparison to the fully nonlinear dynamic analysis method. Thus, inelastic design procedure, using equivalent linear method, cannot adequately guarantee the structural safety for mid-rise building frames resting on soft soils. However, results obtained from the fully nonlinear method of analysis fit the experimental results reasonably well. Therefore, this method is recommended to be used by practicing engineers.

• Coupled systems mechanics
• /
• v.12 no.4
• /
• pp.297-313
• /
• 2023

This paper deals with delamination analysis of non-linear viscoelastic multilayered beam subjected to bending in the plane of the layers. For this purpose, first, a non-linear viscoelastic model is presented. In order to take into account the non-linear viscoelastic behaviour, a non-linear spring and a non-linear dashpot are assembled in series with a linear spring connected in parallel to a linear dashpot. The behaviours of the non-linear spring and dashpot are described by applying non-linear stress-strain and stress-rate of strain relationships, respectively. The constitutive law of the model is derived. Due to the non-linear spring and dashpot, the constitutive law is non-linear. This law is used for describing the time-dependent mechanical behaviour of the beam under consideration. The material properties involved in the constitutive law vary along the beam length due to the continuous material inhomogeneity of the layers. Solution of the strain energy release rate for the delamination is obtained by analyzing the balance of the energy with considering of the non-linear viscoelastic behaviour. The strain energy release rate is found also by using the complementary strain energy for verification. A parametric study is carried-out by using the solution obtained. The solutions derived and the results obtained help to understand the time-dependent delamination of non-linear viscoelastic beams under loading in the plane of layers.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.21 no.6
• /
• pp.562-569
• /
• 2011

In this study linear and nonlinear dynamic stability characteristics of a medium-size high-speed turbocharger, whose rotor is supported by two 3-lobe journal bearings, are analyzed to evaluate and identify the effects of its bearing design variables. The rotor has the rated speed of 40,500 rpm and maximum continuous speed of 45,000 rpm. At first, utilizing the linear stability analysis method, bearing designs of yielding stable or unstable LogDecs as small as possible are searched by manipulating with machined bearing clearances and preloads. As next, utilizing the nonlinear analysis method, limit cycles of the rotor responses at the rated and maximum continuous speeds are simulated to check their acceptances. Results have shown that for the turbocharger rotor-bearing system considered, the 3-lobe journal bearing design with a smaller machined clearance and a larger preload are preferred for the stable rotor responses. More importantly, since there exists a good correlation between the linear and nonlinear stability analysis results, it is concluded that firstly the linear stability analysis method may be applied to screen quickly the ranges of bearing designs for stable or least unstable solutions and then, lastly the nonlinear stability analysis method may be deployed to check an absolute motion stability in terms of the limit cycle.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.5
• /
• pp.477-482
• /
• 2015

Linear regressions and evolutionary nonlinear regression based compensation techniques for the short-range prediction of wind speed are investigated. Development of an efficient MOS(Model Output Statistics) is necessary to correct systematic errors of the model, but a linear regression based MOS is hard to manage an irregular nature of weather prediction. In order to solve the problem, a nonlinear and symbolic regression method using GP(Genetic Programming) is suggested for a development of MOS for wind speed prediction. The proposed method is compared to various linear regression methods for prediction of wind speed. Also, statistical analysis of distribution for UM elements for each method is executed. experiments are performed for KLAPS(Korea Local Analysis and Prediction System) re-analysis data from 2007 to 2013 year for Jeju Island and Busan area in South Korea.

• Interaction and multiscale mechanics
• /
• v.5 no.1
• /
• pp.27-40
• /
• 2012

The two-dimensional incompressible flow of a linear viscoelastic fluid we considered in this research has rapidly oscillating initial conditions which contain both the large scale and small scale information. In order to grasp this double-scale phenomenon of the complex flow, a multiscale analysis method is developed based on the mathematical homogenization theory. For the incompressible flow of a linear viscoelastic Maxwell fluid, a well-posed multiscale system, including averaged equations and cell problems, is derived by employing the appropriate multiple scale asymptotic expansions to approximate the velocity, pressure and stress fields. And then, this multiscale system is solved numerically using the pseudospectral algorithm based on a time-splitting semi-implicit influence matrix method. The comparisons between the multiscale solutions and the direct numerical simulations demonstrate that the multiscale model not only captures large scale features accurately, but also reflects kinetic interactions between the large and small scale of the incompressible flow of a linear viscoelastic fluid.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10a
• /
• pp.362-367
• /
• 1996

In this paper, the eigenstructure of a class of linear time varying systems, termed as linear quasi-time invariant(LQTI) systems, is investigated. A system composed of dynamic devices such as linear time varying capacitors and resistors can be an example of the class. To effectively describe and analyze the LQTI systems, a generalized differential operator G is introduced. Then the dynamic systems described by the operator G are studied in terms of eigenvalue, frequency characteristics, stability and an extended convolution. Some basic attributes of the operator G are compared with those of the differential operator D. Also the corresponding generalized Laplace transform pair is defined and relevant properties are derived for frequency domain analysis of the systems under consideration. As an application example, a LQTI circuit is examined by using the concept of eigenstructure of LQTI system. The LQTI filter processes the sinusoidal signals modulated by some functions.