A simplified forging simulation tool: validation with finite element method

Abstract: Nowadays, the forging industry has a great importance in the fabrication of metallic parts. Therefore, new theoretical models, or modifications to others already existing, continuously appear to improve the process. New investigations culminated, recently, in equations designed to obtain the necessary pressure to forge a part. The purpose is to have a fast, not expensive and efficient alternative to the numerical methods in such calculation. With the double intention of divulgating this new theory created in the University of Málaga (Spain) and confirming its validity, it is interesting to carry out a comparison with the finite element method. In this comparison, geometrical factors relative to the workpiece (part subjected to forging) and the friction coefficient come into play. The goal is then to find validity ranges in the equations for the variables previously mentioned, with the aim to delimit those forging situations in which they could be applied. The implementation of these variables into the commercial FE-code Abaqus is simple, although two aspects arise: the meshing and the correct modeling of the contact between the workpiece and the upper die. The limitation of the Abaqus Student Version (used throughout the analyses) in the allowed number of nodes, prevented performing sufficiently dense meshes. Many alternatives are investigated that, unfortunately, do not show any satisfactory result for the simulation of the process. Performing forging simulations is quite complex and requires a great knowledge in the field and an extraordinary knowledge of how different elements work in large plastic strain. However, this project leaves a collecting data method and a procedure to determine the initial geometry of the workpiece to ensure plastification and to reach a determined shape factor after the forging. The most important is that the way to further analyses has been cleared by confirming the suitability for certain parameters in the simulation (element type, non-linear geometry, contact model, etc.). The recommendation for future simulation efforts is to use ALE (Arbitrary Lagrangian-Eulerian) adaptive meshing.