Premium
Identification of elastoplastic microscopic material parameters within a homogenization scheme
Author(s) -
Schmidt U.,
Mergheim J.,
Steinmann P.
Publication year - 2015
Publication title -
international journal for numerical methods in engineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.421
H-Index - 168
eISSN - 1097-0207
pISSN - 0029-5981
DOI - 10.1002/nme.4933
Subject(s) - homogenization (climate) , identification (biology) , convergence (economics) , scale (ratio) , parameter identification problem , computer science , system identification , identification scheme , mathematics , finite element method , scheme (mathematics) , scale parameter , mathematical optimization , biological system , algorithm , model parameter , mathematical analysis , physics , engineering , structural engineering , data mining , measure (data warehouse) , statistics , ecology , botany , quantum mechanics , economics , economic growth , biodiversity , biology
Summary A two‐scale parameter identification approach is investigated. The microscopic material parameters of a two‐scale model are identified by comparing macroscopic simulation data with macroscopic full‐field measurements of the micro‐structured specimen. Gradient‐based solution strategies are employed for the optimization problem of the two‐scale parameter identification. In particular, two approaches for the gradient calculation are investigated: the finite difference method is compared with a newly introduced semi‐analytical scheme. The focus lies on the identification of microscopic elastoplastic material parameters. The presented identification example with artificial data confirms a reduced computational effort and advantageous convergence for the semi‐analytical approach within the two‐scale parameter identification. A drawback is the increase in memory requirement. Copyright © 2015 John Wiley & Sons, Ltd.