Non standard samples behaviour law parameters determination by inverse analysis

Matthieu Petitprez (1) , Katia Mocellin (1)

1 CEMEF - Centre de Mise en Forme des Matériaux

Electrical contact crimping is a process commonly used in aeronautical and aero spatial applications. One require-ment to perform a resistant assembly is to master crimping parameters. Numerical simulation is then an efficient tool to limit tedious experimental test campaigns. Nevertheless, the numerical model accuracy depends on input data determina-tion, such as material behaviour law parameters. This paper deals with study done to determine material behaviour law parameters of component involved on electri-cal crimping process. The two samples, a strand and an electrical contact, have both small dimensions and specific shapes. For that reason, normalised tensile or compression tests could not be perform. Indeed, inverse analysis is a powerful tool to determine the behaviour laws parameters. The goal is to precisely model the sample mechanical solicitation. A series of computations is launched with various rheological parameters. By comparing experimental and simulated results, the goal is to minimize a cost function. Thus, the first step of the study has been to determine experimental tests to perform in order to gather the force/displacement experimental data. The copper strands have been studied in compression, with a micro indentation de-vice. A 60N force sensor has been used to acquire force data, whereas an inductive displacement sensor has been used for displacement acquisition data. The 60N force sensor is too weak to crush completely a copper contact. For that sample, we decided to use an instrumented crimping plier. A series of crimping tests has been performed on empty barrel, and the equivalent simulation has been done. With the best copper strand material parameters set, relative errors between experimental and simulated copper strand data finally amount to 6.9% (lower than the 8% experimental data scattering). On the other side, with the best cop-per contact material parameters set, the cost function amounts to 4% (lower than the 7% experimental data scattering). This study allows to determine material parameters sets which are use to simulate efficiently the crimping process.

Mots clés: crimping; finite element computations; inverse analysis; automatic optimization

Domaines: Matériaux

Matthieu Petitprez, Katia Mocellin. Non standard samples behaviour law parameters determination by inverse analysis. Computer Methods in Materials Science, 2013, 13 (1), pp.56-62. ⟨hal-01092672⟩