Blank optimization in a stamping process-Influence of the geometry definition

  • Authors:

  • R. De-Carvalho;S. Silva;R. A. F. Valente;A. Andrade-Campos

  • Affiliations:

  • Department of Mechanical Engineering, Centre for Mechanical Technology and Automation, GRIDS Research Group, University of Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal;Department of Mechanical Engineering, Centre for Mechanical Technology and Automation, GRIDS Research Group, University of Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal;Department of Mechanical Engineering, Centre for Mechanical Technology and Automation, GRIDS Research Group, University of Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal;Department of Mechanical Engineering, Centre for Mechanical Technology and Automation, GRIDS Research Group, University of Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal

  • Venue:
  • Finite Elements in Analysis and Design
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

Nowadays initial geometry optimization methods are increasingly being adopted in order to solve complex mechanical plastic forming processes. This kind of approach can focus, in estimating the initial shape of a certain metallic specimen (or blank) in order to achieve a desired geometry after forming. In the present work the superplastic forming of a carter is described and studied in detail. After plastic forming it is possible to verify an undesirable non-homogeneous thickness distribution in the final geometry. A non-uniform thickness distribution of the initial blank can be proposed in order to obtain a regular final thickness of the sheet and avoid this problem. In the present paper the blank surface is modeled by means of a non-uniform rational B-spline (NURBS) surface, where the coordinates of specific NURBS control net vertices are chosen to be the optimization variables. The optimization procedure is carried out by combining a finite element (FE) software and a suitable optimization code. Four different studies were performed, differing in the number of control vertices that formulates the NURBS surface (16, 25, 36 and 49 control vertices were considered), in order to study the influence of the initial geometry definition. A geometry definition leading to better results is achieved, considering both computational cost and final result precision, and subsequent discussions are carried out.