Skip to Main content Skip to Navigation

Optimisation des procédés d'emboutissage par caractérisation géométrique et essais numériques

Abstract : Stamping operation is a process of sheet metal forming. In general, several stamping stages are necessary and the phenomena caused by the plastic strain are complex. So, the design of new tools is a long and delicate operation. The technique of "trial-error" is usually used in the industries. This method is without guarantee of success, expensive and mobilizes a production machine (press). This thesis proposes an optimization based on numerical simulations. The starting point is the definition of the initial stamping tools. Its simulation is carried out. According to the obtained numerical geometry, we parameterize the defects to be correct. The parameters of the process which modify the geometry of the part and their respective range of variation are selected. An experimental design is used in the aim to test different configurations of the tool. The whole simulated experiments is carried out, then, the geometrical parameters are measured. These results permit to calculate polynomial relations between process parameters and geometrical parameters of the part. By using these relations, we seek, by the Newton optimization technique, the optimal configuration of the process giving the desired geometry. Then, the corresponding simulation is carried out and validates the found optimum. Three applications are presented. Two concern parts stamped in one stage, and finally, we present the study of a part obtained in several stages.
Document type :
Complete list of metadatas

Cited literature [40 references]  Display  Hide  Download
Contributor : Yann Ledoux <>
Submitted on : Wednesday, September 23, 2009 - 11:36:35 AM
Last modification on : Monday, October 19, 2020 - 11:13:06 AM
Long-term archiving on: : Wednesday, June 16, 2010 - 12:02:20 AM


  • HAL Id : tel-00419320, version 1



Yann Ledoux. Optimisation des procédés d'emboutissage par caractérisation géométrique et essais numériques. Sciences de l'ingénieur [physics]. Université de Savoie, 2005. Français. ⟨tel-00419320⟩



Record views


Files downloads