Modeling of recrystallization in cold rolled copper using inverse cellular automata and genetic algorithms

Abstract

The process of static recrystallization in annealed and cold rolled Cu was simulated using cellular automata (CA) based model whose parameters were genetically evolved. Before simulating the recrystallization, the initial microstructure, i.e., that of the annealed and cold rolled copper was generated using a simple inverse CA coupled to differential evolution (DE), a real coded variant of genetic algorithms. Dislocation density profiles were assigned to the CA cells of the initial microstructure as per the experimental observations on the variation of nano-hardness within the grains. Each CA cell was assigned a nucleation probability based on the dislocation density corresponding to it. The migration of grain boundaries of the recrystallized grains also depend on the dislocation density distribution. Recovery phenomenon was incorporated the computing scheme by decreasing the dislocation density with time, based on the experimental data on the variation in hardness with annealing time at temperatures below the recrystallization temperature. After formulating the CA model for static recrystallization, DE was used to search for the values of nucleation rate and a constant factor in the growth rate, so that a match between the simulated and the experimentally observed microstructures was achieved.