| Multi-stage forming processes are used to deform billets of simple shape to products with complex geometries and are characterized by multitudes of process and design parameters. For optimal selection of design parameters for such processes, an efficient integrated analysis-design method is necessary that will minimize human intervention while maintaining a high level of accuracy in optimality considerations. In the recent past, biologically inspired Genetic Algorithms (GAs) optimization technique based on probabilistic transition rules, have been successfully implemented for a wide variety of problems in physical and social sciences, engineering, manufacturing and operations research, and computer science. Micro Genetic Algorithms ($mu$GA) have evolved to reduce the large computation time required for Simple Genetic Algorithms (SGA) based optimization schemes. In this study, a design tool has been developed for multi-stage metal forming processes based on the coupling of Micro Genetic Algorithms for optimal design with finite element analysis of metal forming processes using general purpose packages. The main advantages of this technique are its modular nature and its capability to handle a large number of discrete and continuous design variables.;Three different multi-stage metal forming processes have been considered to demonstrate the applicability of the developed technique. They are: (i) Multi-pass cold wire drawing process where the objectives are, to obtain uniform distribution of plastic strains in the drawn wire, and to minimize the total deformation energy for the process; (ii) Multi-pass profile drawing where the objective is to minimize underfilling to ensure dimensional accuracy at critical regions of a drawn profile; and (iii) Cold forging of an automotive outer race preform where the objective is to minimize the failure by tensile cracking of the cold forged product. General purpose finite element analysis packages NIKE2D, DEFORM3D and DEFORM2D have been used for evaluation of objective functions for the three processes respectively. Application of Micro Genetic Algorithms based design optimization technique results in significant improvement in the distribution of product properties, in the reduction of deformation energy and in the reduction of the number of forming stages for the processes considered. |
