CAE-based application for identification and verification of hyperelastic parameters

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

The main objective of this study is to develop a CAE-based application with a convenient GUI for identification and verification of material parameters for hyperelastic models available in the current release of the FE code ANSYS Mechanical APDL. This Windows application implements a two-step procedure: (1) fitting of experimental stress–strain curves provided by the user; (2) verification of the obtained material parameters by the solution of a modified benchmark problem. The application, which was developed using the Visual Basic.NET language, implements a two-way interaction with ANSYS as a single loop using the APDL script as input and text, graphical and video files as output. With this application, nine isotropic incompressible hyperelastic material models are compared by fitting them to the conventional Treloar’s experimental dataset (1944) for vulcanised rubber. A ranking of hyperelastic models is constructed according to model efficiency, which is estimated using fitting quality criteria. The model ranking is done based upon the complexity of their mathematical formulation and their ability to accurately reproduce the test data. Recent hyperelastic models (Extended Tube and Response Function) are found to be more efficient compared to conventional ones. The verification is done by the comparison of an analytical solution to an FEA result for the benchmark problem of a rubber cylinder under compression proposed by Lindley (1967). In the application, the classical formulation of the benchmark is improved mathematically to become valid for larger deformations. The wide applicability of the proposed two-step approach is confirmed using stress–strains curves for seven different formulations of natural rubber and seven different grades of synthetic rubber.
LanguageEnglish
Pages611-626
Number of pages16
JournalProceedings of the Institution of Mechanical Engineers, Part L: Journal of Materials: Design and Applications
Volume231
Issue number7
Early online date1 Sep 2015
DOIs
Publication statusPublished - 1 Oct 2017

Fingerprint

Computer aided engineering
Rubber
ANSYS
Stress-strain curves
Benchmark
Formulation
Ranking
Model
Elastomers
Synthetic rubber
Hyperelastic Material
Curve
Response Function
Large Deformation
Graphical user interfaces
Identification (control systems)
Tube
Analytical Solution
Compression
Valid

Keywords

  • benchmark
  • elastomers
  • finite element analysis (FEA)
  • hyperelasticity
  • ranking
  • software

Cite this

@article{214232ca5c52411f8a2eb7e0b0b6f4cc,
title = "CAE-based application for identification and verification of hyperelastic parameters",
abstract = "The main objective of this study is to develop a CAE-based application with a convenient GUI for identification and verification of material parameters for hyperelastic models available in the current release of the FE code ANSYS Mechanical APDL. This Windows application implements a two-step procedure: (1) fitting of experimental stress–strain curves provided by the user; (2) verification of the obtained material parameters by the solution of a modified benchmark problem. The application, which was developed using the Visual Basic.NET language, implements a two-way interaction with ANSYS as a single loop using the APDL script as input and text, graphical and video files as output. With this application, nine isotropic incompressible hyperelastic material models are compared by fitting them to the conventional Treloar’s experimental dataset (1944) for vulcanised rubber. A ranking of hyperelastic models is constructed according to model efficiency, which is estimated using fitting quality criteria. The model ranking is done based upon the complexity of their mathematical formulation and their ability to accurately reproduce the test data. Recent hyperelastic models (Extended Tube and Response Function) are found to be more efficient compared to conventional ones. The verification is done by the comparison of an analytical solution to an FEA result for the benchmark problem of a rubber cylinder under compression proposed by Lindley (1967). In the application, the classical formulation of the benchmark is improved mathematically to become valid for larger deformations. The wide applicability of the proposed two-step approach is confirmed using stress–strains curves for seven different formulations of natural rubber and seven different grades of synthetic rubber.",
keywords = "benchmark, elastomers, finite element analysis (FEA), hyperelasticity, ranking, software",
author = "Yevgen Gorash and Tugrul Comlekci and Robert Hamilton",
year = "2017",
month = "10",
day = "1",
doi = "10.1177/1464420715604004",
language = "English",
volume = "231",
pages = "611--626",
journal = "Proceedings of the Institution of Mechanical Engineers, Part L: Journal of Materials: Design and Applications",
issn = "1464-4207",
number = "7",

}

TY - JOUR

T1 - CAE-based application for identification and verification of hyperelastic parameters

AU - Gorash, Yevgen

AU - Comlekci, Tugrul

AU - Hamilton, Robert

PY - 2017/10/1

Y1 - 2017/10/1

N2 - The main objective of this study is to develop a CAE-based application with a convenient GUI for identification and verification of material parameters for hyperelastic models available in the current release of the FE code ANSYS Mechanical APDL. This Windows application implements a two-step procedure: (1) fitting of experimental stress–strain curves provided by the user; (2) verification of the obtained material parameters by the solution of a modified benchmark problem. The application, which was developed using the Visual Basic.NET language, implements a two-way interaction with ANSYS as a single loop using the APDL script as input and text, graphical and video files as output. With this application, nine isotropic incompressible hyperelastic material models are compared by fitting them to the conventional Treloar’s experimental dataset (1944) for vulcanised rubber. A ranking of hyperelastic models is constructed according to model efficiency, which is estimated using fitting quality criteria. The model ranking is done based upon the complexity of their mathematical formulation and their ability to accurately reproduce the test data. Recent hyperelastic models (Extended Tube and Response Function) are found to be more efficient compared to conventional ones. The verification is done by the comparison of an analytical solution to an FEA result for the benchmark problem of a rubber cylinder under compression proposed by Lindley (1967). In the application, the classical formulation of the benchmark is improved mathematically to become valid for larger deformations. The wide applicability of the proposed two-step approach is confirmed using stress–strains curves for seven different formulations of natural rubber and seven different grades of synthetic rubber.

AB - The main objective of this study is to develop a CAE-based application with a convenient GUI for identification and verification of material parameters for hyperelastic models available in the current release of the FE code ANSYS Mechanical APDL. This Windows application implements a two-step procedure: (1) fitting of experimental stress–strain curves provided by the user; (2) verification of the obtained material parameters by the solution of a modified benchmark problem. The application, which was developed using the Visual Basic.NET language, implements a two-way interaction with ANSYS as a single loop using the APDL script as input and text, graphical and video files as output. With this application, nine isotropic incompressible hyperelastic material models are compared by fitting them to the conventional Treloar’s experimental dataset (1944) for vulcanised rubber. A ranking of hyperelastic models is constructed according to model efficiency, which is estimated using fitting quality criteria. The model ranking is done based upon the complexity of their mathematical formulation and their ability to accurately reproduce the test data. Recent hyperelastic models (Extended Tube and Response Function) are found to be more efficient compared to conventional ones. The verification is done by the comparison of an analytical solution to an FEA result for the benchmark problem of a rubber cylinder under compression proposed by Lindley (1967). In the application, the classical formulation of the benchmark is improved mathematically to become valid for larger deformations. The wide applicability of the proposed two-step approach is confirmed using stress–strains curves for seven different formulations of natural rubber and seven different grades of synthetic rubber.

KW - benchmark

KW - elastomers

KW - finite element analysis (FEA)

KW - hyperelasticity

KW - ranking

KW - software

U2 - 10.1177/1464420715604004

DO - 10.1177/1464420715604004

M3 - Article

VL - 231

SP - 611

EP - 626

JO - Proceedings of the Institution of Mechanical Engineers, Part L: Journal of Materials: Design and Applications

T2 - Proceedings of the Institution of Mechanical Engineers, Part L: Journal of Materials: Design and Applications

JF - Proceedings of the Institution of Mechanical Engineers, Part L: Journal of Materials: Design and Applications

SN - 1464-4207

IS - 7

ER -