Analyzing the Fluid Flow in Continuous Casting through Evolutionary Neural Nets and Multi-Objective Genetic Algorithms

The flow fields computed for a typical continuous caster are analysed using the basic concepts of Pareto-optimality in the context of multi-objective optimization. The data generated by the flow solver FLUENT (TM) are trained through Evolutionary Neural Networks that emerged through a Pareto-tradeoff between the complexity of the network and its accuracy of training. A number of objectives constructed this way are subjected to optimization using a Multi-objective Predator-Prey Genetic Algorithm. The procedure is repeated using the software mode-FRONTIER (TM) and the results are compared and analysed.