Abstract
An overview of modern optimization methods, including Evolutionary Al-gorithms (EAs) and gradient{based optimization methods adapted for Cluster and Grid Computing is presented. The basic tool is a Hierarchical Distributed Metamodel{Assisted EA supporting Multilevel Evaluation, Multilevel Search and Multilevel Parameterization. In this framework, the adjoint method computes the first and second derivatives of the objective function with respect to the design variables, for use in aerodynamic shape optimization. Such a multi{component, hierarchical and distributed scheme requires particular attention when Cluster or Grid Computing is used and a much more delicate parallelization compared to that of conventional EAs. © 2009 Springer-Verlag Berlin Heidelberg.
Cite
CITATION STYLE
Giannakoglou, K. C., Kampolis, I. C., Liakopoulos, P. I. K., Karakasis, M. K., Papadimitriou, D. I., Zervogiannis, T., & Asouti, V. G. (2009). Aerodynamic shape optimization methods on multiprocessor platforms. In Lecture Notes in Computational Science and Engineering (Vol. 67 LNCSE, pp. 49–58). https://doi.org/10.1007/978-3-540-92744-0_6
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
