Premium
Identification of Parameters on an Interpolated Function with Measurement Errors
Author(s) -
Nostitz Niklas,
Hendrik Kröger Nils,
Ihlemann Jörn
Publication year - 2018
Publication title -
pamm
Language(s) - English
Resource type - Journals
ISSN - 1617-7061
DOI - 10.1002/pamm.201800337
Subject(s) - identification (biology) , interpolation (computer graphics) , levenberg–marquardt algorithm , computer science , function (biology) , algorithm , basis (linear algebra) , finite element method , mathematical optimization , optimization algorithm , mathematics , artificial intelligence , engineering , artificial neural network , motion (physics) , geometry , structural engineering , biology , botany , evolutionary biology
While the identification of material parameters via FEM is an important basis for efficient simulations, the identification itself leads to new problems. In general, each optimization step and error function evaluation is gained with high computational effort. To speed up the identification, an alternative interpolation algorithm is introduced with a gradient free optimization. The combination of this algorithm with a Levenberg‐Marquardt algorithm improves the results with uncertain data.