The Kriging-based genetic algorithm is applied to aerodynamic design problems. Kriging model, one of the response surface models, represents a relationship between objective function (output) and variables (input) using stochastic process. kriging model drastically reduces computational time required for evaluation in optimization (optimum searching) ‘Expected improvement (EI)’ used as criterion select additional sample points. This makes it possible not only improve accuracy but also...

Analysis of variance (ANOVA) and self-organizing map (SOM) were applied to data mining for aerodynamic design space. These methods make it possible identify the effect each variable on objective functions. ANOVA shows information quantitatively, while SOM qualitatively. Furthermore, can show effects interaction between variables functions visualize trade-offs among This will be helpful designers determine final from non-dominated solutions multi-objective problems. two results: a fly-back...

In this study, a surrogate model is applied to multi-objective aerodynamic optimization design. For the balanced exploration and exploitation with model, objective functions are converted Expected Improvements (EI) these values directly used as fitness in optimization. Among non-dominated solutions about EIs, additional sample points for update of Kriging selected. The present method transonic airfoil order obtain information design space, two data mining techniques results. One analysis...

A sophisticated GA/PSO-hybrid algorithm for application to real-world optimization problems was proposed. The configurations of the two consisting methods, GA and PSO, were investigated enhance diversity former fast convergence latter simultaneously. new hybrid applied test function problems, results indicated that search ability improved by suitable tuning configurations. In addition, showed robust regardless selection initial population. also a diesel engine combustion chamber design...

Diesel engine combustion chamber which reduces exhaust emission has been designed using CFD analysis and optimization techniques. In order to save computational time for design, the Kriging model, one of response surface models, is adopted here. For a robust exploration, both estimated function value model its uncertainty are considered at same time. present problem, k-means method used limit number additional sample points reasonable level. Among points, two shapes dominate baseline...

A new approach, Multi-Objective Design Exploration (MODE), is presented to address Multidisciplinary Optimization problems. MODE reveals the structure of design space from trade-off information and visualizes it as a panorama for Decision Maker. The present form consists Kriging Model, Adaptive Range Multi Objective Genetic Algorithms, Analysis Variance Self-Organizing Map. main emphasis this approach visual data mining. Two mining examples using high fidelity simulation codes are presented:...

*† ‡ § ¶ # In this paper, a practical Multidisciplinary Design Optimization (MDO) system for an aircraft design is developed. The MDO based on the integration of computational fluid dynamics (CFD) codes and NASTRAN aeroelastic-structural interface code. Kriging model employed to save time objective function evaluation in Multi-Objective Genetic Algorithm (MOGA). As result optimization, several nondominated solutions, indicating trade-off among drag, structural weight, drag divergence...

This paper presents a comparison of the criteria for updating Kriging surrogate models in multi-objective optimization: expected improvement (EI), hypervolume (EHVI), estimation (EST), and those combination (EHVI + EST). EI has been conventionally used as criterion considering stochastic each objective function value individually, while EHVI recently proposed front nondominated solutions optimization. EST is estimated nonstochastically by model without its uncertainties. Numerical...

A multi-objective design exploration for a three-element airfoil consisting of slat, main wing, and flap was carried out. The lift curve improvement is important to high-lift system, thus has be performed with considered multi-angle. objective functions here are maximize the coefficient at landing near-stall conditions simultaneously. Genetic algorithm used as an optimizer. Although it advantage global exploration, its computational cost expensive. To reduce cost, Kriging surrogate model,...

A helicopter rotor is optimally designed for aeroacoustic performance improvement. As shown in previous reports, the blade shapes can be to minimize high-speed impulsive noise but tend have excessively high tapers and swept back. Since an overly short chord length around blade-tip region may cause structural problems safety issues autorotation, autorotation index has been introduced keep tip from having excessive taper ratios. In addition, changes thickness camber of airfoils also taken into...

This paper presents a comparison of the criteria for updating Kriging surrogate models in surrogate-based non-constrained many-objective optimization: expected improvement (EI), hypervolume (EHVI), and estimate (EST). EI has been conventionally used as criterion considering stochastic each objective function value individually, while EHVI proposed front nondominated solutions multi-objective optimization. EST is estimated non-stochastically by model without its uncertainties. Numerical tests...

This article presents a combined use of multi-objective optimization and quantitative design rule mining methods to improve the aerodynamic efficiency stability centrifugal impeller with vaned diffuser. A time-averaged but spatially distributed flow is considered at mixing plane evaluate uniformity, which affects stability. First, impeller's shape has been optimized using genetic algorithm uniformity. It was found that trade-off among non-dominated solutions can be controlled by vane-less...

One of the difficulties in multi-disciplinary design optimization lies complicated interactions between large numbers objective functions, variables, and constraints. This difficulty often leads to an unsuitable formulation problems. Data mining is used address these challenges. provides insight into systems. The information obtained from data can be support (a) problems, (b) decision making, (c) steering. report presents a review recent developments applications techniques engineering...

The data mining has been performed for the aerodynamic design optimization result of two-stage-to-orbit reusable launch vehicle flyback booster wing. Three techniques were used such as self-organizing map, functional analysis variance, and rough set theory. problem had four objective functions 71 variables regarding wing shape. obtained hypothetical database with 302 all solutions including 102 non-dominated solutions. Consequently, knowledge in space was acquired correlation between...

A new approach for multi-objective robust design optimization was proposed and applied to a practical problem with large number of objective functions. The present is assisted by response surface approximation visual data-mining, resulted in two major gains regarding computational time data interpretation. Kriging model can markedly reduce the predictions robustness. In addition, use self-organizing maps as data-mining technique allows visualization complicated information between optimality...

Genetic Algorithms (GAs) generally maintain diverse solutions of good quality in multi-objective problems, while Particle Swarm Optimization (PSO) shows rapid convergence to the optimum solution. Previous studies indicated that search abilities can be improved by simply coupling these two algorithms; GA compensates for low diversity PSO, PSO high computational costs GA. In this study, configurations methods when used a fully coupled hybrid algorithm were investigated achieve an improvement...

This paper describes the design optimization of a sport shoe sole structure by evolutionary computation coupled with eigenmode analysis based on finite element method. A genetic algorithm assisted Kriging response surface model was used for global and efficient optimization. The present study implemented two cases equivalent problem formulations: single-objective constrained multi-objective non-constrained problem. described here provided midsole optimal material properties, in which elastic...

This paper compares the criteria for updating Kriging response surface models in multi-objective optimization: expected improvement (EI), hypervolume (EHVI), estimation (EST), and those combination (EHVI+EST). EI has been conventionally used as criterion considering stochastic of each objective function value individually, while EHVI recently proposed front non-dominated solutions optimization. EST is function, which estimated non-stochastically by model without its uncertainties. Numerical...

The problem of designing a centrifugal blower was explored using multi-objective genetic algorithm and data mining techniques. Blade-to-blade regions an impeller diffuser were modeled time-averaged non-uniform inflow to the considered. design objectives efficiency uniformity diffuser. impeller’s shape represented by NURBS curves then optimized. obtained non-dominated solutions showed trade-off relationship variables controlling found be related dimensions vane-less load balance impeller. We...

Multi-objective design optimization for a steam turbine stator blade was implemented using three-dimensional large eddy simulation (LES) and genetic algorithm (GA). The GA used here assisted by the Kriging response surface model global efficient optimization. aim of described to reduce overall pressure loss local due end walls simultaneously. results revealed candidates that overcame baseline in terms loss, trade-off relation between them. In addition, these provided specific concept...