• Composites Research
• /
• v.16 no.4
• /
• pp.36-43
• /
• 2003

In this research, the optimization of composite structures was performed using metamodels. The optimization of composite structures requires a lot of time when optimizing the result of the time-consuming analysis. Thus, metamodels are used to replace the time-consuming analysis with simple models. RSM, kriging and neural networks are widely used metamodels. RSM and kriging were used in this study. The ultimate failure load analysis of the composite structure was approximated by metamodels. The optimizations of the composite plate were performed to maximize ultimate failure load using genetic algorithm and metamodels.

• Journal of computational fluids engineering
• /
• v.14 no.4
• /
• pp.31-40
• /
• 2009

Although optimization by sequentially refining metamodels is known to be computationally very efficient, the metamodel that can be used for this purpose is limited to Kriging method due to the difficulties related with sample points selections. The present study suggests a novel method for sequentially refining metamodels using Pareto Frontiers, which can be used independent of the type of metamodels. It is shown from the examples that the present method yields more accurate metamodels compared with full-factorial optimization and also guarantees global optimum irrespective of the initial conditions. Finally, in order to prove the generality of the present method, it is applied to a 2D transonic airfoil optimization problem, and the successful design results are obtained.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.27 no.3
• /
• pp.381-386
• /
• 2003

Gradually engineering designers are determined based on computer simulations. Modeling of the computer simulation however is too expensive and time consuming in a complicate system. Thus, designers often use approximation models called metamodels, which represent approximately the relations between design and response variables. There arc general metamodels such as response surface model and kriging metamodel. Response surface model is easy to obtain and provides explicit function. but it is not suitable for highly nonlinear and large scaled problems. For complicate case, we may use kriging model that employs an interpolation scheme developed in the fields of spatial statistics and geostatistics. This class of into interpolating model has flexibility to model response data with multiple local extreme. In this study. metamodeling techniques are adopted to carry out the shape optimization of a funnel of Cathode Ray Tube. which finds the shape minimizing the local maximum principal stress Optimum designs using two metamodels are compared and proper metamodel is recommended based on this research.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.34 no.5
• /
• pp.541-547
• /
• 2010

The rigorous validation of the accuracy of metamodels is an important topic in research on metamodel techniques. Although a leave-k-out cross-validation technique involves a considerably high computational cost, it cannot be used to measure the fidelity of metamodels. Recently, the mean$_0$ validation technique has been proposed to quantitatively determine the accuracy of metamodels. However, the use of mean$_0$ validation criterion may lead to premature termination of a sampling process even if the kriging model is inaccurate. In this study, we propose a new validation technique based on the mean and variance of the response evaluated when sequential sampling method, such as maximum entropy sampling, is used. The proposed validation technique is more efficient and accurate than the leave-k-out cross-validation technique, because instead of performing numerical integration, the kriging model is explicitly integrated to accurately evaluate the mean and variance of the response evaluated. The error in the proposed validation technique resembles a root mean squared error, thus it can be used to determine a stop criterion for sequential sampling of metamodels.

• Journal of Navigation and Port Research
• /
• v.34 no.7
• /
• pp.553-558
• /
• 2010

This study presents the optimization design process of a segment ball valve that involves the reduction of the flow resistance coefficient and the satisfaction of the strength requirement. Numerical analysis of fluid flow and structural analysis have been performed to predict the flow resistance coefficient and the maximum stress of a segment ball valve. In this study, a segment ball valve incorporating the advantages of a ball valve and a butterfly valve has been devised. In general, ball valves are installed in a pipe system where tight shut off is required. Butterfly valves having smaller end-to-end dimension than ball valve can be installed in narrow spaces in a pipe system. The metamodels for the shape design of a segment ball valve are built by the response surface method and the Kriging interpolation model.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.30 no.6 s.249
• /
• pp.691-697
• /
• 2006

Metamodel, model of model, has been widely used to improve an efficiency of optimization process in engineering fields. However, global metamodels of constraints in a constrained optimization problem are required good accuracy around neighborhood of optimum point. To satisfy this requirement, more sampling points must be located around the boundary and inside of feasible region. Therefore, a new sampling strategy that is capable of identifying feasible domain should be applied to select sampling points for metamodels of constraints. In this research, we suggeste sequential feasible domain sampling that can locate sampling points likely within feasible domain by using penalty function method. To validate the excellence of feasible domain sampling, we compare the optimum results from the proposed method with those form conventional global space-filling sampling for a variety of optimization problems. The advantages of the feasible domain sampling are discussed further.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.19 no.5
• /
• pp.27-37
• /
• 2020

Structural optimization problems with discrete design variables require more function calculations (or finite element analyses) than those in the continuous design space. In this study, a method to find an optimal solution in the discrete design of the truss structure is presented, reducing the number of function calculations. Because a continuous optimal solution is the Karush-Kuhn-Tucker point that satisfies the optimality condition, it is assumed that the discrete optimal solution is around the continuous optimum. Then, response values such as weight, displacement, and stress are predicted using approximate models-referred to as hybrid metamodels-within specified design ranges. The discrete design method using the hybrid metamodels is used as a post-process of the continuous optimization process. Standard truss design problems of 10-bar, 25-bar, 15-bar, and 52-bar are solved to show the usefulness of this method. The results are compared with those of existing methods.

• Proceedings of the KSME Conference
• /
• 2008.11a
• /
• pp.1169-1175
• /
• 2008

The validation technique is classified with two methods whether to demand of additional experimental points. The method which requires additional experimental points such as RSME is actually impossible in engineering field. Therefore, the method which only use experimented points such as the cross validation technique is only available. But the cross validation not only requires considerable computational costs for generating metamodel each iterations, but also cannot measure quantitatively the fidelity of metamodel. In this research we propose a new validation technique for representative metamodels using an variance of metamodel and confidence interval information. The proposed validation technique computes confidence intervals using a variance information from the metamodel. This technique will have influence on choosing the accurate metamodel, constructing ensemble of each metamodels and advancing effectively sequential sampling technique.

• Journal of the military operations research society of Korea
• /
• v.25 no.2
• /
• pp.1-14
• /
• 1999

Metamodels using response surface methodology (RSM) are used for the optimality analysis of linear programming (LP). They have the form of a simple polynomial, and predict the optimal objective function value of an LP for various levels of the constraints. The metamodelling techniques for optimality analysis of LP can be applied to large-scale LP models. What is needed is some large-scale application of the techniques to verify how accurate they are. In this paper, we plan to use the large scale LP model, strategic transport optimal routing model (STORM). The developed metamodels of the large scale LP can provide some useful information.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.34 no.8
• /
• pp.989-996
• /
• 2010

Statistical regression and/or interpolation models have been used for data analysis and response prediction using the results of the physical experiments and/or computer simulations in structural engineering fields. These models have been employed during the last decade to develop a variety of design methodologies. However, these models only handled responses with respect to space variables such as size and shape of structures and cannot handle time-variant dynamic responses, i.e. response varying with time. In this research, statistical space-time metamodels based on multiple response approach that can handle responses with respect to both space variables and a time variable are proposed. Regression and interpolation models such as the response surface model (RSM) and kriging model were developed for handling time-variant dynamic responses of structural engineering. We evaluate the accuracies of the responses predicted by the two statistical space-time metamodels by comparing them with the responses obtained by the physical experiments and/or computer simulations.