• Communications for Statistical Applications and Methods
• /
• v.26 no.1
• /
• pp.69-78
• /
• 2019

A structured model with both single-index and varying coefficients is a powerful tool in modeling high dimensional data. It has been widely used because the single-index can overcome the curse of dimensionality and varying coefficients can allow nonlinear interaction effects in the model. For high dimensional index vectors, variable selection becomes an important question in the model building process. In this paper, we propose an efficient estimation and a variable selection method based on a smoothing spline approach in a partially linear single-index-coefficient regression model. We also propose an efficient algorithm for simultaneously estimating the coefficient functions in a data-adaptive lower-dimensional approximation space and selecting significant variables in the index with the adaptive LASSO penalty. The empirical performance of the proposed method is illustrated with simulated and real data examples.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.19 no.3
• /
• pp.713-723
• /
• 1995

In intelligent machine tools, a computer based control system, which can adapt the machining parameters in an optimal fashion based on sensor measurements of the machining process, should be incorporated. In this paper, the technology for adaptively optimizing the cutting conditions to maximize the material removal rate in face milling operations is proposed using the exterior penalty function method combined with multilayered neural networks. Two neural networks are introduced ; one for estimating tool were length, the other for mapping input and output relations from experimental data. Then, the optimization of cutting conditions is adaptively implemented using tool were information and predicted process output. The results are demonstrated with respect to each level of machining such as rough, fine and finish cutting.

• Proceedings of the KIEE Conference
• /
• 1998.07c
• /
• pp.1111-1113
• /
• 1998

A robust economic dispatch algorithm involving transmission losses is proposed and investigated for a possibility of on-line applications. In this paper, the penalty factors are calculated directly from transposed Jacobian of load flow analysis with advantages of superiority to B-coefficients method based on its computation time and suitability for real time application since the approach is based on a current system condition. The proposed algorithm is systematically handling the generation capacity constraints with transmission losses. Implementation of the algorithm for IEEE systems and EPRI Scenario systems shows that computation time is enough to apply on-line economic dispatch to large power system and production cost is saved compared with the crude classical economic dispatch algorithm without considering transmission losses.

• Structural Engineering and Mechanics
• /
• v.62 no.4
• /
• pp.391-399
• /
• 2017

This study investigates the effects of soft story, short columns, heavy overhangs, pounding, and construction and workmanship quality parameters on seismic response of reinforced concrete buildings through nonlinear static and dynamic procedures. The accounted parameters are selected for their common use in rapid screening of RC buildings. The 4- and 7-story buildings designed according to pre-modern codes are used to reflect majority of the existing building stock. The relative penalty scores are employed in this study to evaluate relative importance of certain irregularities in the existing rapid seismic assessment procedures. Comparison of relative scores for the irregularities considered in this study show that the overall trend is similar. The relatively small differences may be accounted for regional construction practices. It is concluded that initial-phase seismic assessment procedures based on architectural features yield in somewhat similar results independent of their bases. However, the differences in the scores emphasize the proper selection of the method based on the regional structure characteristics.

• Communications for Statistical Applications and Methods
• /
• v.29 no.3
• /
• pp.319-331
• /
• 2022

In this paper, we develop a new statistical model to forecast the PM_2.5 level in Seoul, South Korea. The proposed model is based on the extreme quantile regression model with lasso penalty. Various meteorological variables and air pollution variables are considered as predictors in the regression model, and the lasso quantile regression performs variable selection and solves the multicollinearity problem. The final prediction model is obtained by combining various extreme lasso quantile regression estimators and we construct a binary classifier based on the model. Prediction performance is evaluated through the statistical measures of the performance of a binary classification test. We observe that the proposed method works better compared to the other classification methods, and predicts 'very bad' cases of the PM_2.5 level well.

• Journal of Theoretical and Applied Mechanics
• /
• v.1 no.1
• /
• pp.97-109
• /
• 1995

In this paper, a finite viscoelastic continuum model for rubber and its finite element analysis are presented. This finite viscoelatic model based on continuum mechanics is an extended model of Johnson and Wuigley's 1-D model. In this extended model, continuum based kinematic measures are rigorously defied and by using this kinematic measures, elastic stage law and flow rule are introduced. In kinematics, three configuration are introduced. In kinematics, three configuration are introduced. They are reference, current and virtual visco configurations. In elastic state law, it is assumed that at a certain time, there exists an elastic potential which describes the recoverable elastic energy. From this elastic potential, elastic state law is derived. The proposed flow rule is based on phenomenological observation. The flow rule gives precise relaxation response. In finite element approximation, mixed Lagrangian description is used, where total and similar method of updated Lagrangian descriptions are used together. This approach reduces the numerical job and gives simple nonlinear syatem of equations. To satisfy the incompressible condition, penalty-type modified Mooney-Rivlin energy function is adopted. By this method nearly incompressible condition is obtain the virtual visco configuration. For verification, uniaxial stretch tests are simulated for various stretch rates. The simulated results show good agreement with experiments. As a practical experiments. As a preactical example, pressurized rubber plate is simulated. The result shows finite viscoelastic effects clearly.

• Coupled systems mechanics
• /
• v.2 no.4
• /
• pp.389-410
• /
• 2013

Three-dimensional Lagrangian fluid finite element is applied to seismic response analysis of an oil storage tank with a floating roof. The fluid element utilized in the present analysis is formulated based on the displacement finite element method considering only volumetric elasticity and its element stiffness matrix is derived by using one-point integration method in order to avoid volumetric locking. The method usually adds a rotational penalty stiffness to satisfy the irrotational condition for fluid motion and modifies element mass matrices through the projected mass method to suppress spurious hourglass-mode appeared in compensation for one-point integration. In the fluid element utilized in the present paper, a small hourglass stiffness is employed. The fluid and structure domains for the objective oil storage tank are modeled by eight-node solid elements and four-node shell elements, respectively, and the transient response of the floating roof structure or the free surface are evaluated by implicit direct time integration method. The results of seismic response analyses are compared with those by other method and the validation of the present analysis using three-dimensional Lagrangian fluid finite elements is shown.

• Journal of the Society of Naval Architects of Korea
• /
• v.51 no.4
• /
• pp.328-333
• /
• 2014

A hybrid particle-mesh method based on the vorticity-velocity formulation for solving the incompressible Navier-Stokes equations is a combination of the Vortex-In-Cell(VIC) method for convection and the penalization method for diffusion. The key feature of the numerical methods is to determine velocity and vorticity fields around a solid body on a temporary grid, and then the time evolution of the flow is computed by tracing the convection of each vortex element using the Lagrangian approach. Assuming that the vorticity and velocity fields are to be computed in time domain analysis, pressure fields are estimated through a complete set of solutions at present time step. It is possible to obtain vorticity and velocity fields prior to any pressure calculation since the pressure term is eliminated in the vorticity-velocity formulation. Therefore, pressure field is explicitly treated by solving a suitable Poisson equation. In this paper, we propose a simple way to numerically implement the vorticity-velocity-pressure formulation including a penalty term. For validation of the proposed numerical scheme, we illustrate the early development of viscous flows around an impulsive started circular cylinder for Reynolds number of 9500.

• Structural Engineering and Mechanics
• /
• v.27 no.4
• /
• pp.409-424
• /
• 2007

In this paper, an identification method of impact force is proposed for composite structures. In this method, the relation between force histories and strain responses is first formulated. The transfer matrix, which relates the strain responses of sensors and impact force information, is constructed from the finite element method (FEM). Based on this relation, an optimization model to minimize the difference between the measured strain responses and numerically evaluated strain responses is built up to obtain the impact force history. The identification of force history is performed by a modified least-squares method that imposes the penalty on the first-order derivative of the force history. Moreover, from the relation of strain responses and force history, an error vector indicating the force location is defined and used for the force location identification. The above theory has also been extended into the cases when using acceleration information instead of strain information. The validity of the present method has been verified through two experimental examples. The obtained results demonstrate that the present approach works very well, even when the internal damages in composites happen due to impact events. Moreover, this method can be used for the real-time health monitoring of composite structures.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.49 no.8
• /
• pp.423-431
• /
• 2000

This paper presents a systematic approach to the input-output data-based fuzzy modeling for the complex and uncertain nonlinear systems, in which the conventional mathematical models may fail to give the satisfying results. To do this, we propose a new method that can yield a successful fuzzy model using a mGA hybrid schemes with a fine-tuning method. We also propose a new coding method fo chromosome for applying the mGA to the structure and parameter identifications of fuzzy model simultaneously. During mGA search, multi-purpose fitness function with a penalty process is proposed and adapted to guarantee the accurate and valid fuzzy modes. This coding scheme can effectively represent the zero-order Takagi-Sugeno fuzzy model. The proposed mGA hybrid schemes can coarsely optimize the structure and the parameters of the fuzzy inference system, and then fine tune the identified fuzzy model by using the gradient descent method. In order to demonstrate the superiority and efficiency of the proposed scheme, we finally show its applications to two nonlinear systems.