• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2004.10a
• /
• pp.437-444
• /
• 2004

In this study, an optimum support design problem is considered to minimize displacement of stacked plates under self weight condition. During the displacement analysis, several kinds of contact arise between the plates themselves and support bar. These can be easily considered if commercial analysis software, which provides capability to solve the contact problem, is used. It is found, however, that the computing time is extraordinarily long due possibly to the generality of the software and also to the ignorance of the control parameters used in the software. In this paper, the contact condition is imposed directly by the authors, while the software is used only to solve the ordinary displacement analysis problem. In this way, the computing time is decreased remarkably by more than 30 times, while yielding the same accurate results. Optimization is conducted based on this efficient analysis method to find minimum number of supporting bars using the response surface algorithm.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.1
• /
• pp.227-235
• /
• 2017

When the spatial information of each location is given specifically as coordinates it is popular to use the geographically weighted regression to incorporate the spatial information by assuming that the regression parameters vary spatially across locations. In this paper, we relax the linearity assumption of geographically weighted regression and propose a geographically weighted least squares-support vector machine for estimating geographically weighted mean by using the basic concept of kernel machines. Generalized cross validation function is induced for the model selection. Numerical studies with real datasets have been conducted to compare the performance of proposed method with other methods for predicting geographically weighted mean.

• Journal of Institute of Control, Robotics and Systems
• /
• v.11 no.7
• /
• pp.565-571
• /
• 2005

We present a novel speed estimation of a Permanent Magnet Synchronous Motor(PMSM) based on Support Vector Regression(SVR). The proposed method can estimate wide speed range, including 0.33Hz with full load, accurately in the steady and transient states where motor parameters variations are known without parameter estimator. Moreover, the method does not need offline training previously but is trained on-line. The training starts with the PMSM operation simultaneously and estimates the speed in real time. The experimental results shows the validity and the usefulness of the proposed algorithm for the 0.4Kw PMSM DSP(TMS320VC33) drive system.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2008.04a
• /
• pp.603-607
• /
• 2008

This paper is to investigate the effects of support conditions on the whirling stability of a supercritical composite driveshaft. Two system parameters are rigorously considered: one is the bending stiffness of the support beam/bearings and the other is the rotating internal damping of the shaft. An analytic model is developed based on finite element methods and an eigenvalue analysis is employed to estimate the shaft stability under supercritical operating conditions. Results show that the internal damping causes the whirling instability at a supercritical speed, as demonstrated in other previous studies. However, the bending stiffness of the support beam is found to affect greatly the stability behaviors of a supercritical shaft and several combinations of the shaft/beam stiffness can be identified to guarantee the stable operation even in a supercritical region.

• Structural Engineering and Mechanics
• /
• v.61 no.3
• /
• pp.335-345
• /
• 2017

Free vibration analysis for continuous bridge under any number of vehicles is conducted in this paper. Calculation strategy for natural frequency and mode shape is proposed based on Euler-Bernoulli beam theory and numerical assembly method. Firstly, a half-car planar model is adopted; equations of motion and displacement functions for bridge and vehicle are established, respectively. Secondly, the undermined coefficient matrices for wheels, vehicles, intermediate support, left-end support and right-end support are derived. Then, the numerical assembly technique for conventional finite element method is adopted to construct the overall matrix of coefficients for whole system. Finally, natural frequencies and corresponding mode shapes are determined based on iterative method and overall matrix solution. Numerical simulation is presented to verify the effectiveness of the proposed method. The results reveal that the solutions of present method are exact ones. Natural frequencies and associate modal shapes of continuous bridge under different conditions of vehicles are investigated. The influences of vehicle parameters on natural frequencies are also demonstrated.

• Journal of the Korean Chemical Society
• /
• v.58 no.6
• /
• pp.543-552
• /
• 2014

A quantitative structure activity relationship (QSAR) study is performed for modeling and prediction of oral bioavailabilities of 216 diverse set of drugs. After calculation and screening of molecular descriptors, linear and nonlinear models were developed by using multiple linear regression (MLR), artificial neural network (ANN), support vector machine (SVM) and random forest (RF) techniques. Comparison between statistical parameters of these models indicates the suitability of SVM over other models. The root mean square errors of SVM model were 5.933 and 4.934 for training and test sets, respectively. Robustness and reliability of the developed SVM model was evaluated by performing of leave many out cross validation test, which produces the statistic of $Q^2_{SVM}=0.603$ and SPRESS = 7.902. Moreover, the chemical applicability domains of model were determined via leverage approach. The results of this study revealed the applicability of QSAR approach by using SVM in prediction of oral bioavailability of drugs.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.30 no.1
• /
• pp.75-94
• /
• 2005

Research in the decision sciences has continued to develop a variety of mathematical models as well as software tools supporting corporate decision-making. Yet. in spite of their potential usefulness, the models are little used in real-world decision making since the model solution processes are too complex for ordinary users to get accustomed. This paper proposes an intelligent and flexible model-solver integration framework that enables the user to solve decision problems using multiple models and solvers without having precise knowledge of the model-solution processes. Specifically, for intuitive model-solution, the framework enables a decision support system to suggest the compatible solvers of a model autonomously without direct user intervention and to solve the model by matching the model and solver parameters intelligently without any serious conflicts. Thus, the framework would improve the productivity of institutional model solving tasks by relieving the user from the burden of leaning model and solver semantics requiring considerable time and efforts.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2004.11a
• /
• pp.179-188
• /
• 2004

Bankruptcy prediction is an important and widely studied topic since it can have significant impact on bank lending decisions and profitability. Recently, support vector machine (SVM) has been applied to the problem of bankruptcy prediction. The SVM-based method has been compared with other methods such as neural network, logistic regression and has shown good results. Genetic algorithm (GA) has been increasingly applied in conjunction with other AI techniques such as neural network, CBR. However, few studies have dealt with integration of GA and SVM, though there is a great potential for useful applications in this area. This study proposes the methods for improving SVM performance in two aspects: feature subset selection and parameter optimization. GA is used to optimize both feature subset and parameters of SVM simultaneously for bankruptcy prediction.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.55 no.7
• /
• pp.306-312
• /
• 2006

Support Vector Machine(SVM), of which the foundations have been developed by Vapnik (1995), is gaining popularity thanks to many attractive features and promising empirical performance. In this paper, we propose a new short-term load forecasting technique based on SVM. We discuss the input vector selection of SVM for load forecasting and analyze the prediction performance for various SVM parameters such as kernel function, cost coefficient C, and $\varepsilon$ (the width of 8 $\varepsilon-tube$). The computer simulation shows that the prediction performance of the proposed method is superior to that of the conventional neural networks.

• Structural Engineering and Mechanics
• /
• v.91 no.2
• /
• pp.197-209
• /
• 2024

The vibration of elastically supported bidirectional functionally graded (BDFG) sandwich beams on an elastic foundation is investigated. The sandwich structure is composed of upper and lower layers of BDFG material and the core layer of isotropic material. Material properties of upper and lower layers are assumed to vary continuously along the length and thickness of the beam with a power-law function. Hamilton's principle is used to deduce the vibration equations of motion of the sandwich Timoshenko beam. Then, the partial differential equation of motion is spatially discretized into a time-varying ordinary differential equation in terms of Chebyshev differential matrices. The eigenvalue equation associated with the free vibration is formulated to study the influence of various slenderness ratios, material gradient indexes, thickness ratios, foundation and support spring constants on the vibration frequency of BDFG sandwich beams. The present method can provide researchers with deep insight into the impact of various geometric, material, foundation and support parameters on the vibration behavior of BDFG sandwich beam structures.