Variable-fidelity aerodynamic shape optimization

Abstract

Aerodynamic shape optimization (ASO) plays an important role in the design of aircraft, turbomachinery and other fluid machinery. Simulation-driven ASO involves the coupling of computational fluid dynamics (CFD) solvers with numerical optimization methods. Although being relatively mature and widely used, ASO is still being improved and numerous challenges remain. This chapter provides an overview of simulation-driven ASO methods, with an emphasis on surrogate-based optimization (SBO) techniques. In SBO, a computationally cheap surrogate model is used in lieu of an accurate high-fidelity CFD simulation in the optimization process. Here, a particular focus is given to SBO exploiting surrogate models constructed from corrected physics-based low-fidelity models, often referred to as variable-or multi-fidelity optimization. © 2011 Springer-Verlag Berlin Heidelberg.