Inverse analysis for identification of parameters during thermo-mechanical tests

Abstract : Sensitivity matrix evaluation in gradient method for inverse analysis is a major of importance. The techniques vary from the optimisation problems, the main feature of the direct models and the type of applications. The semi-analytical method of sensitivity analysis of inverse problems is very attractive thanks to the compromise between computational time and ease of derivatives evaluation. Especially for parameter identification in material forming domain, this technique seems to be promising. This paper presents a comparison between semi-analytical and finite difference methods. The framework of this work is the rheological and tribological parameter identification through material forming process analysis. The identification software CART (Computer Aided Rheology and Tribology) was developed from the finite element code FORGE2(R) devoted to material forming processes simulation.