Modelling of the post-localization behaviour in tube hydroforming of low carbon steels
Autor: | P. Duroux, Jean-Marc Roelandt, J. Wilsius, D. Daly, Mohamed Rachik |
---|---|
Rok vydání: | 2007 |
Předmět: |
Hydroforming
Materials science Computer simulation business.industry Metals and Alloys Structural engineering Strain rate Instability Industrial and Manufacturing Engineering Finite element method Computer Science Applications Modeling and Simulation Ultimate tensile strength Ceramics and Composites Formability Process simulation business |
Zdroj: | Journal of Materials Processing Technology. 182:248-256 |
ISSN: | 0924-0136 |
Popis: | In tube hydroforming, some local thinning can be observed on industrial components. When low carbon steels are concerned, the process is stopped broadly after reaching the plastic instability (strain localization) while having an acceptable part. In numerical simulation, the classical elastic–plastic models reach their limits when the plastic instability occurs, the process simulation becomes very sensitive to numerical parameters which leads to poor prediction of post-localization behaviour. Consequently, a full understanding of post-localization is important for the formability prediction purpose. In this work, various numerical simulations were carried out on both tensile and hydroforming tests. The classical Prandtl–Reuss model is first investigated and the obtained results clearly show that this model is not appropriate to predict the post-localization behaviour. Then, some regularization techniques such as material rate dependence and the introduction of gradient-based internal variable in the constitutive relations were evaluated. In the light of the obtained results, the rate-dependent model seems to be well adapted to predict the post-localization behaviour in tube hydroforming simulation. Thus, the rate-dependent approach was adopted and the basic model was enhanced to fit some experimental data from tensile tests. Finally, the hydroforming Corner Fill test was simulated and the numerical results were compared with measurements to validate the proposed approach. This comparison shows that the rate-dependent model greatly improves the accuracy of the predictions. |
Databáze: | OpenAIRE |
Externí odkaz: |