Surrogate-Assisted Design Optimization Using Response Features

Slawomir Koziel and Leifur Leifsson
Additional contact information
Slawomir Koziel: Reykjavik University, Engineering Optimization & Modeling Center
Leifur Leifsson: Iowa State University, Department of Aerospace Engineering

Chapter Chapter 9 in Simulation-Driven Design by Knowledge-Based Response Correction Techniques, 2016, pp 147-163 from Springer

Abstract: Abstract Highly nonlinear system responses, whether the nonlinearity is with respect to the adjustable parameters of the systems under design or with respect to certain free parameters (e.g., chord line (Leifsson and Koziel 2015a, b, c), time, or frequency (Koziel et al. 2006a, b)), are challenging to simulation-driven design optimization. The surrogate-based optimization techniques described in the previous chapters directly handled the relevant responses of the systems of interest (such as aerodynamic forces in case of aerodynamic shape optimization problems, or S-parameters versus frequency for microwave/photonic and antenna devices). One of the issues is the limited generalization capability of the surrogate models which may lead to a degradation of the convergence properties of the optimization algorithms. Fortunately, in many cases, it is possible to reformulate the optimization problem in terms of the so-called response features whose dependence on the optimization variables is much less nonlinear than for the original responses. Feature-based optimization executed at the level of these feature (or characteristic) points may be computationally much more efficient, even when realized using single-fidelity simulation models only. In this chapter, we formulate feature-based optimization, demonstrate its application for the design of microwave/photonic and antenna devices, as well as discuss its limitations and generalizations.

Keywords: Feature Point; Microring Resonator; Fractional Bandwidth; Pattern Search Algorithm; Antenna Device (search for similar items in EconPapers)
Date: 2016
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-319-30115-0_9

Ordering information: This item can be ordered from
http://www.springer.com/9783319301150

DOI: 10.1007/978-3-319-30115-0_9

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().