Design of optimal aerodynamic shapes using stochastic optimization methods and computational intelligence

Optimization and inverse design methods based on stochastic and, in particular, evolutionary algorithms, are currently being used in aeronautics since they offer a useful tool to help solving current and future technological problems. Despite their advantages, all of the population-based search algorithms require excessive CPU time due to the excessive number of candidate solutions which need to be evaluated through costly computational models. This paper focuses on the reduction of this computing cost, so as to make stochastic optimization both efficient and effective. Emphasis is laid on techniques which rely on the construction and use of surrogate or approximation models which may substitute for the exact and costly evaluation tool. A literature survey of a number of relevant methods is followed by numerous indicative examples, which demonstrate the usefulness of these methods. (C) 2002 Elsevier Science Ltd. All rights reserved. (C) 2002 Elsevier Science Ltd. All rights reserved.