Response surface approximations for structural optimization

Abstract

Response surface methodology can be used to construct global and midrange approximations to functions in structural optimization. Since structural optimization requires expensive function evaluations, it is important to construct accurate function approximations so that rapid convergence may be achieved. In this paper techniques to ÿnd the region of interest containing the optimal design, and techniques for ÿnding more accurate approximations are reviewed and investigated. Aspects considered are experimental design techniques, the selection of the 'best' regression equation, intermediate response functions and the location and size of the region of interest. Standard examples in structural optimization are used to show that the accuracy is largely dependent on the choice of the approximating function with its associated subregion size, while the selection of a larger number of points is not necessarily cost-e ective. In a further attempt to improve e ciency, di erent regression models were investigated. The results indicate that the use of the two methods investigated does not signiÿcantly improve the results. Finding an accurate global approximation is challenging, and su cient accuracy could only be achieved in the example problems by considering a smaller region of the design space. ? 1998 John Wiley & Sons, Ltd.

Comparison of ultra capacitors, hydraulic accumulators and batteries technologies to optimize hybrid vehicle ecoefficiency

“…Because of the larger number of function evaluations that can be necessary to carry out optimization process, especially when using meta heuristic algorithms, a standard approach in structural and multidisciplinary optimization consists in resorting to global or local approximation models (see for instance [14], [15]). Approximations will replace direct simulation runs during optimization iterations and will be updated during a limited number of steps ( [16]).…”

Evaluation de la vulnérabilité sismique des constructions en béton armé en vue de leur réhabilitation

“…02003-p.5 Les propriétés matérielles qui ont été utilisées pour le béton sont : la résistance à la compression qui varie selon le tableau 2, résistance à la traction A partir des résultats du plan factoriel complet présentés dans le tableau 3, il est possible de développer un modèle de régression polynomiale définissant des surfaces de réponses [12]. Ces modèles ne sont valables que dans le domaine des paramètres qui ont été utilisés pour les établir et leurs extrapolations à des valeurs ne se trouvant pas dans ce domaine devraient être soumises à des épreuves de validation.…”

FEA-Based Optimization of Blank Holder Force and Pressure for Hydromechanical Deep Drawing of Parabolic Cup using Greedy Search and RSM Methods

“…Roux et al [6] discussed experimental design techniques and regression equations for structural optimization. Response surface methodology combining with stochastic finite elements were used by Kleiber et al [7] for reliability assessment in metal forming.…”