Parallel evolutionary computation for solving complex CFD optimization problems: a review and some nozzle applications

Abstract

This paper attempts to present a summarized survey in the field of Parallel Evolutionary Algorithms (PEAs), trying to highlight the most relevent aspects in the design and implementation of each class of PEAs. A review of the most advanced research activities on this field is presented New tools for new challenges of Industry and Society, which are definitely multi criteria oriented and CFD dominant in the applications, are investigated and a series of simple nozzle shape optimization problems with their methodologies and associated results - including Hierarchy and Game Theory - are discussed Data resulting from evolutionary optimization, which have been collected in the database of a European Thematic Network named INGEnet are presented as a road map to Multidisciplinary Design Optimisation.