• International Journal of Highway Engineering
• /
• v.14 no.5
• /
• pp.189-199
• /
• 2012

PURPOSES: The purpose of this study is to propose a new methodology for developing statistical collision models and to show the validation results of the methodology. METHODS: A new modeling method of introducing variables into the model one by one in a multiplicative form is suggested. A method for choosing explanatory variables to be introduced into the model is explained. A method for determining functional forms for each explanatory variable is introduced as well as a parameter estimating procedure. A model selection method is also dealt with. Finally, the validation results is provided to demonstrate the efficacy of the final models developed using the method suggested in this study. RESULTS: According to the results of the validation for the total and injury collisions, the predictive powers of the models developed using the method suggested in this study were better than those of generalized linear models for the same data. CONCLUSIONS: Using the methodology suggested in this study, we could develop better statistical collision models having better predictive powers. This was because the methodology enabled us to find the relationships between dependant variable and each explanatory variable individually and to find the functional forms for the relationships which can be more likely non-linear.

• The Korean Journal of Applied Statistics
• /
• v.19 no.3
• /
• pp.407-419
• /
• 2006

In this paper, we analyze household overdue loans in Korea which has been causing serious social and economical problems. We consider customers of Bank A in Korea and focus on overdue cash services which have been snowballing in the past few years. From analysis of overdue loans, one can predict possible delays for current customers as well as build a credit evaluation and risk management system for future customers. As a statistical analytical tool, we propose a two-stage Generalized Linear regression Model (GLM) which assumes a logistic model for presence/non-presence of overdue and a gamma model for the amount of overdue in the case of overdue. We perform goodness of fit test for the two-stage model and select significant explanatory variables in each stage of the model. It turns out that age, the amount of credit loans from other financial companies, the amount of cash service from other companies, debit balance, the average amount of cash service, and net profit are important explanatory variables relevant to overdue credit card cash service in Korea.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.24 no.5 s.176
• /
• pp.1075-1083
• /
• 2000

The prediction of the inelastic behavior of the structure is an essential part of reliability assessment procedure, because most of the failures are induced by the inelastic deformation, such as creep and plastic deformation. During decades, there has been much progress in understanding of the inelastic behavior of the materials and a lot of inelastic constitutive equations have been developed. These equations consist of the definition of inelastic strain and the evolution of the state variables introduced to quantify the irreversible processes occurred in the material. With respect to the definition of the inelastic strain, the inelastic constitutive models can be categorized into elastoplastic model, unified viscoplastic model and separated viscoplastic model and the different integration methods have been applied to each category. In the present investigation, the generalized integration method applicable for various types of constitutive equations is developed and implemented into ABAQUS by means of UMAT subroutine. The solution of the non-linear system of algebraic equations arising from time discretization with the generalized midpoint rule is determined using line-search technique in combination with Newton method. The strategy to control the time increment for the improvement of the accuracy of the numerical integration is proposed. Several numerical examples are considered to demonstrate the efficiency and applicability of the present method. The prediction of the inelastic behavior of the structure is an essential part of reliability assessment procedure, because most of the failures are induced by the inelastic deformation, such as creep and plastic deformation. During decades, there has been much progress in understanding of the inelastic behavior of the materials and a lot of inelastic constitutive equations have been developed. These equations consist of the definition of inelastic strain and the evolution of the state variables introduced to quantify the irreversible processes occurred in the material. With respect to the definition of the inelastic strain, the inelastic constitutive models can be categorized into elastoplastic model, unified viscoplastic model and separated viscoplastic model and the different integration methods have been applied to each category. In the present investigation, the generalized integration method applicable for various types of constitutive equations is developed and implemented into ABAQUS by means of UMAT subroutine. The solution of the non-linear system of algebraic equations arising from time discretization with the generalized midpoint rule is determined using line-search technique in combination with Newton method. The strategy to control the time increment for the improvement of the accuracy of the numerical integration is proposed. Several numerical examples are considered to demonstrate the efficiency and applicability of the present method.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.6
• /
• pp.2354-2363
• /
• 2018

This paper proposed a novel double closed control strategy for single-phase voltage source pulse width modulation (PWM) rectifier based on active disturbance rejection control (ADRC) and dq current decoupling control. First, the mathematical model of the single-phase PWM rectifier in the d-q axis synchronous rotating reference frame is established by constructing a virtual component using a second-order generalized integrator (SOGI). Then, the mathematical model is simplified according to the active power conservation, and the first-order equation of single-phase PWM rectifier voltage outer loop is acquired. A linear auto-disturbance rejection controller is used to design the voltage outer loop according to the first-order equation. Finally, the proposed control strategy and the traditional PI control are compared and verified by simulation and physical experiments. Both simulation and experimental results confirm that the proposed control strategy has excellent dynamic performance and strong rejection ability to disturbances.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.26 no.8
• /
• pp.1167-1177
• /
• 1989

In this paper, a generalized adaptive law is proposed which uses a rational function type operator for parameter adjustment. To satisfy the passivity condition of the adaptation block, we introduce a constant feedback gain into the adaptation block. This adaptation scheme is applied to the model reference adaptive control of a continuous-time, linear time-invariant, minimum-phase system whose relative degree is 1. We prove the asymptotic stability of the output error of this adaptive system by hyperstability method. It is shown that by digital computer simulations this law can give a better output error transient response in some cases than the conventional gradient adaptive law. And the output error responses for the several types of the proposed adaptation law are examined in the presence of a kind of unmodeled dynamics.

• Communications for Statistical Applications and Methods
• /
• v.16 no.4
• /
• pp.697-705
• /
• 2009

The Pearson chi-squared statistic or the deviance statistic is widely used in assessing the goodness-of-fit of the generalized linear models. But these statistics are not proper in the situation of continuous explanatory variables which results in the sparseness of cell frequencies. We propose a goodness-of-fit test statistic for the cumulative logit models with ordinal responses. We consider the grouping of a dataset based on the ordinal scores obtained by fitting the assumed model. We propose the Pearson chi-squared type test statistic, which is obtained from the cross-classified table formed by the subgroups of ordinal scores and the response categories. Because the limiting distribution of the chi-squared type statistic is intractable we suggest the parametric bootstrap testing procedure to approximate the distribution of the proposed test statistic.

• Journal of applied mathematics & informatics
• /
• v.14 no.1_2
• /
• pp.193-212
• /
• 2004

A new model of generalized thermoelasticity equations for isotropic media with temperature-dependent mechanical properties is established. The modulus of elasticity is taken as a linear function of reference temperature. The present model is described both generalizations, Lord Shulman (L-S) theory with one relaxation time and Green-Lindsay (G-L) with two relaxation times, as well as the coupled theory, instantaneously. The method of the matrix exponential, which constitutes the basis of the state space approach of modern control theory, applied to two-dimensional equations. Laplace and Fourier integral transforms are used. The resulting formulation is applied to a problem of a thick plate subject to heating on parts of the upper and lower surfaces of the plate that varies exponentially with time. Numerical results are given and illustrated graphically for the problem considered. A comparison was made with the results obtained in case of temperature-independent modulus of elasticity in each theory.

• Computers and Concrete
• /
• v.7 no.4
• /
• pp.365-386
• /
• 2010

This paper presents the physical formulation of a 1D material model suitable for seismic applications. It is written within the framework of thermodynamics with internal variables that is, especially, very efficient for the phenomenological representation of material behaviors at macroscale: those of the representative elementary volume. The model can reproduce the main characteristics observed for concrete, that is nonsymetric loading rate-dependent (viscoelasticity) behavior with appearance of permanent deformations and local hysteresis (continuum plasticity), stiffness degradation (continuum damage), cracking due to displacement localization (discrete plasticity or damage). The parameters have a clear physical meaning and can thus be easily identified. Although this point is not detailed in the paper, this material model is developed to be implemented in a finite element computer program. Therefore, for the benefit of the robustness of the numerical implementation, (i) linear state equations (no local iteration required) are defined whenever possible and (ii) the conditions in which the presented model can enter the generalized standard materials class - whose elements benefit from good global and local stability properties - are clearly established. To illustrate the capabilities of this model - among them for Earthquake Engineering applications - results of some numerical applications are presented.

• Structural Engineering and Mechanics
• /
• v.65 no.4
• /
• pp.369-380
• /
• 2018

In this paper, a multi-layered finite element model for laminated glass plates is introduced. A layer-wise theory is applied to the analysis of laminated glass due to the combination of stiff and soft layers; the independent layers are connected via Lagrange multipliers. The von $K{\acute{a}}rm{\acute{a}}n$ large deflection plate theory and the constant Poisson ratio for constitutive equations are assumed to capture the possible effects of geometric nonlinearity and the time/temperature-dependent response of the plastic foil. The linear viscoelastic behavior of a polymer foil is included by the generalized Maxwell model. The proposed layer-wise model was implemented into the MATLAB code and verified against detailed three-dimensional models in ADINA solver using different hexahedral finite elements. The effects of temperature, load duration, and creep/relaxation are demonstrated by examples.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2018.05a
• /
• pp.157-157
• /
• 2018

A reliable and accurate downscaling model which can provide climate change information, obtained from global climate models (GCMs), at finer resolution has been always of great interest to researchers. In order to achieve this model, linear methods widely have been studied in the past decades. However, nonlinear methods also can be potentially beneficial to solve downscaling problem. Therefore, this study explored the applicability of some nonlinear machine learning techniques such as neural network (NN), extreme learning machine (ELM), and ELM autoencoder (ELM-AE) as well as a linear method, least absolute shrinkage and selection operator (LASSO), to build a reliable temperature downscaling model. ELM is an efficient learning algorithm for generalized single layer feed-forward neural networks (SLFNs). Its excellent training speed and good generalization capability make ELM an efficient solution for SLFNs compared to traditional time-consuming learning methods like back propagation (BP). However, due to its shallow architecture, ELM may not capture all of nonlinear relationships between input features. To address this issue, ELM-AE was tested in the current study for temperature downscaling.