Preform optimal design in metal forging using genetic algorithms

Abstract

A numerical method for shape optimisation in forging is presented. The goal of the optimisation is to eliminate work‐piece defects that may arise during the forging process. A two‐dimensional finite element code has been developed for the simulation of the mechanical process. The material is incompressible and it follows the Norton‐Hoff law. To deal with contact constraint the velocity projection algorithm is used. The optimisation process is conducted using a genetic algorithm supported by an elitist strategy. A new genetic operator called adaptive mutation has been developed to increase the efficiency of the search. The developed scheme is used to design optimal preform shapes for several axisymmetric examples. Continuous and discrete design variables are considered. The objective function of the optimisation problem is associated with the quality of the final product. Comparing the obtained optimal results with the literature validates the proposed optimisation method.

Citation

Poursina, M., António, C.A.C., Castro, C.F., Parvizian, J. and Sousa, L.C. (2004), "Preform optimal design in metal forging using genetic algorithms", Engineering Computations, Vol. 21 No. 6, pp. 631-650. https://doi.org/10.1108/02644400410545209