Non-linear simulation of coiling accounting for roughness of contacts and multiplicative elastic-plastic behavior

Research output: Contribution to journalArticlepeer-review

Abstract

In this paper numerical simulations of coiling (winding of a steel strip on itself) and uncoiling are developed. Initial residual stress field is taken into account as well as roughness of contacts and elastic-plastic behavior at finite strains, considering the Tresca yield function and isotropic hardening. The main output is the residual stress field due to plastic deformations during the process. This enables to quantify additional flatness defects. The presented coiling simulation relies on a modeling strategy that consists in dividing each time step into two sub-steps. Each sub-step can be solved semi-analytically and numerical optimizations enable to obtain a general solution. Thus, reasonable computation times are reached and parametric studies can be performed in order to develop coiling strategies considering the process parameters. Comparisons with previous models from the literature are presented. Moreover, the comparison with a Finite Element simulation presents the same order of magnitude, however, it shows that direct computations using classical FE codes are difficult to perform in terms of computation times and stability if an explicit integration scheme is chosen. Numerical results are also given in order to determine the effect of some parameters such as roughness, yield stress, applied force, strip crown or mandrel's radius.

Original languageEnglish
Pages (from-to)1-20
Number of pages20
JournalInternational Journal of Solids and Structures
Volume94-95
DOIs
Publication statusPublished - 1 Sept 2016
Externally publishedYes

Keywords

  • Coiling
  • Contact
  • Finite strains
  • Multiplicative elasto-plasticity
  • Optimization
  • Roughness

Fingerprint

Dive into the research topics of 'Non-linear simulation of coiling accounting for roughness of contacts and multiplicative elastic-plastic behavior'. Together they form a unique fingerprint.

Cite this