• Proceedings of the KSME Conference
• /
• 2001.06c
• /
• pp.537-542
• /
• 2001

In recent engineering, the designer has become more and more dependent on computer simulation. But defining exact model using computer simulation is too expensive and time consuming in the complicate systems. Thus, designers often use approximation models, which express the relation between design variables and response variables. These models are called metamodel. In this paper, we introduce one of the metamodel, named Kriging. This model employs an interpolation scheme and is developed in the fields of spatial statistics and geostatistics. This class of interpolating model has flexibility to model response data with multiple local extreme. By reason of this multi modality, we can't use any gradient-based optimization algorithm to find global extreme value of this model. Thus we have to introduce global optimization algorithm. To do this, we introduce DE(Differential Evolution). DE algorithm is developed by Ken Price and Rainer Storn, and it has recently proven to be an efficient method for optimizing real-valued multi-modal objective functions. This algorithm is similar to GA(Genetic Algorithm) in populating points, crossing over, and mutating. But it introduces vector concept in populating process. So it is very simple and easy to use. Finally, we show how we determine Kriging metamodel and find global extreme value through two mathematical examples.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.3 no.2
• /
• pp.16-29
• /
• 1995

In this study a tracked vehicle is idealized as a 2-dimensional 9-degrees-of-freedom model which takes into account the effects of HSU units, torsion bars, and track. For the model equations of motion are derived using Kane's method. By using the equations of motion, a numerical example is solved and results are compared to those obtained by using a general purpose multi body dynamic analysis program. The comparison study shows the reasonable coherence between the two results. which confirms the effectiveness of the model. With the model, dynamic response optimization is carried out. The objective function is the peak value of the vertical acceleration of the vehicle at the driver's seat, and the constraints are the wheel travel limits, the ground clearance. and the limits of other design variables. Three different sets of design variables are chosen and used for the optimization. The results show the attenuation of the acceleration peak value. Thus the procedure presented in this study can be utilized for the design improvement of the real system.

• Steel and Composite Structures
• /
• v.51 no.6
• /
• pp.647-659
• /
• 2024

Given the increased interest in enhancing structural sustainability, the current study sought to apply multiobjective optimization to a footbridge with a steel-concrete composite I-girder structure. It was considered as objectives minimizing the cost for building the structure, the environmental impact assessed by CO2 emissions, and the vertical accelerations created by human-induced vibrations, with the goal of ensuring pedestrian comfort. Spans ranging from 15 to 25 meters were investigated. The resistance of the slab's concrete, the thickness of the slab, the dimensions of the welded steel I-profile, and the composite beam interaction degree were all evaluated as design variables. The optimization problem was handled using the Multiobjective Harmony Search (MOHS) metaheuristic algorithm. The optimization results were used to generate a Pareto front for each span, allowing us to assess the correlations between different objectives. By evaluating the values of design variables in relation to different levels of pedestrian comfort, it was identified optimal values that can be employed as a starting point in predimensioning of the type of structure analyzed. Based on the findings analysis, it is possible to highlight the relationship between the structure's cost and CO2 emission objectives, indicating that cost-effective solutions are also environmentally efficient. Pedestrian comfort improvement is especially feasible in smaller spans and from a medium to a maximum level of comfort, but it becomes expensive for larger spans or for increasing comfort from minimum to medium level.

• Korean Institute of Interior Design Journal
• /
• no.29
• /
• pp.86-93
• /
• 2001

This study presents about the method of alternatives selection by considering wall performance and interior design. Wall is selected fur the object and 3 items of cost, performance, and design as the objective function for optimizing are determined. Thus the wall performance selected problems, which are improvement of insulation performance, sweaty prevention, sound insulation performance and design selected problems, which is satisfactory Improvement of users about Interior design. It is important to select alternatives that can satisfy the performance and design on the capital given as much as possible. But quantitative problem such as performance or expanses and qualitative problem such as design are not in the same dimension. Therefore this problem is a multi-criteria optimization problem and also has used AHP method as the method to solve these. Moreover GA is used to solve a problem of the alternatives occurrence, which is the characteristic of multi-criteria problem. This study presents the solution method on multi-criteria problem that has been mix loaded of quantitative problem and qualitative problem by using AHP(Analytic Hierarchy Process) and GA(Genetic Algorithm).

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.19 no.7
• /
• pp.500-511
• /
• 2007

The Optimal solutions of the design variables in internally finned tubes have been obtained for three-dimensional periodically fully developed turbulent flow and heat transfer. For a trapezoidal fin profile, performances of the heat exchanger are determined by considering the heat transfer rate and pressure drop, simultaneously, that are interdependent quantities. Therefore, Pareto frontier sets of a heat exchanger can be acquired by integrating CFD and a multi-objective optimization technique. The optimal values of fin widths $(d_1,\;d_2)$, fin height(h) and helix angle$(\gamma)$ are numerical1y obtained by minimizing the pressure loss and maximizing the heat transfer rate within ranges of $d_1=0.5\sim1.5mm$, $d_2=0.5\sim1.5mm$, $h=0.5\sim1.5mm$, and $\gamma=0\sim20^{\circ}$. For this, a general CFD code and a global genetic algorithm(GA) are used. The Pareto sets of the optimal solutions can be acquired after $30^{th}$ generation.

• Smart Structures and Systems
• /
• v.12 no.5
• /
• pp.523-545
• /
• 2013

Piezoelectric and dielectric behaviors of a piezoceramic patch adhesively centered on a carbon composite plate are identified using a robust multi-objective optimization procedure. For this purpose, the patch piezoelectric stress coupling and blocked dielectric constants are automatically evaluated for a wide frequency range and for the different identifiable behaviors. Latters' symmetry conditions are coded in the design plans serving for response surface methodology-based sensitivity analysis and meta-modeling. The identified constants result from the measured and computed open-circuit frequencies deviations minimization by a genetic algorithm that uses meta-model estimated frequencies. Present investigations show that the bonded piezoceramic patch has effective three-dimensional (3D) orthotropic piezoelectric and dielectric behaviors. Besides, the sensitivity analysis indicates that four constants, from eight, dominate the 3D orthotropic behavior, and that the analyses can be reduced to the electromechanically coupled modes only; therefore, in this case, and if only the dominated parameters are optimized while the others keep their nominal values, the resulting piezoelectric and dielectric behaviors are found to be transverse-isotropic. These results can help designing piezoceramics smart composites for various applications like noise, vibration, shape, and health control.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.12 no.6
• /
• pp.461-468
• /
• 2002

Computer simulation is essential to design the suspension elements of railway vehicle. By computer simulation, engineers can assess the feasibility of the given design factors and change them to get a better design. But if one wishes to perform complex analysis on the simulation, such as railway vehicle dynamic, the computational time can become overwhelming. Therefore, many researchers have used a surrogate model that has a regression model performed on a data sampling of the simulation. In general, metamodels(surrogate model) take the form y($\chi$)＝f($\chi$)＋$\varepsilon$, where y($\chi$) is the true output, f($\chi$) is the metamodel output, and is the error. In this paper, a second order polynomial equation is used as the RSM(response surface model) for high speed train that have twenty-nine design variables and forty-six responses. After the RSM is constructed, multi-objective optimal solutions are achieved by using a nonlinear programming method called VMM(variable matric method) This paper shows that the RSM is a very efficient model to solve the complex optimization problem.

• Smart Structures and Systems
• /
• v.16 no.1
• /
• pp.1-26
• /
• 2015

Minimizing construction cost and reducing seismic damage are two conflicting objectives in the design of any new structure. In the present work, we try to develop a framework in order to solve the optimum performance-based design problem considering the construction cost and the seismic damage of steel moment-frame structures. The Park-Ang damage index is selected as the seismic damage measure because it is one of the most realistic measures of structural damage. The non-dominated sorting genetic algorithm (NSGA-II) is employed as the optimization algorithm to search the Pareto optimal solutions. To improve the time efficiency of the proposed framework, three simplifying strategies are adopted: first, simplified nonlinear modeling investigating minimum level of structural modeling sophistication; second, fitness approximation decreasing the number of fitness function evaluations; third, wavelet decomposition of earthquake record decreasing the number of acceleration points involved in time-history loading. The constraints of the optimization problem are considered in accordance with Federal Emergency Management Agency's (FEMA) recommended seismic design specifications. The results from numerical application of the proposed framework demonstrate the efficiency of the framework in solving the present multi-objective optimization problem.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.42 no.2
• /
• pp.108-118
• /
• 2014

On the optimization design problem using surrogate model, it requires considerable number of sampling points to construct a surrogate model which retains the accuracy. As an alternative to reduce construction cost of the surrogate model, Variable-Fidelity Modeling(VFM) technique, where correct high fidelity model based on the low fidelity surrogate model is introduced. In this study, hierarchical kriging model for variable-fidelity surrogate modeling is used and an optimization framework with multi-objective genetic algorithm(MOGA) is presented. To prove the feasibility of this framework, airfoil design optimization process is performed for the transonic region. The parameters of PARSEC are used to design variables and the optimization process is performed in case of varying number of grid and varying fidelity. The results showed that pareto front of all variable-fidelity models are similar with its single-level of fidelity model and calculation time is considerably reduced. Based on computational results, it is shown that VFM is a more efficient way and has an accuracy as high as that single-level of fidelity model optimization.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.39 no.5
• /
• pp.433-441
• /
• 2011

The optimization of a centrifugal compressor was conducted. The ANN (Artificial Neural Network) was adopted as an optimization algorithm, and it was learned and trained with the DOE (Design of Experiment). In the DOE, it was predicted the main effect and the interaction effect of design variables to the objective function. The ANN was improved in the optimization process using the GA (Genetic Algorithm). When any output at each generation was reached a standard level, it was re-calculated by the CFD (Computational Fluid Dynamics) and it was applied to develop a new ANN. After 6th generation, the prediction difference between ANN and CFD was less than 1%. A pareto of the efficiency versus the pressure ratio was obtained through the 21th generation. Using this method, the computational time for the optimization was equivalent to the time consumed by the gradient method, and the optimized results of multi-objective function were obtained.