Phases in Zn-coated Fe analyzed through an evolutionary meta-model and multi-objective Genetic Algorithms

Various interfaces in Zn coated steel are examined for their stability on the basis of two conflicting requirements of minimum deformation at a maximum absorption of shear energy. The shearing process is simulated using a Molecular Dynamics technique and meta-models of both energy and strain are constructed using an Evolutionary Neural Network that itself evolved through a multi-objective Genetic Algorithm. Simultaneous optimization of deformation and energy absorption is conducted with a Predator-prey Genetic Algorithm and the resulting Pareto frontiers are analyzed and discussed. The findings show good correspondence with existing experimental observations. (C) 2011 Elsevier B.V. All rights reserved.