Finite element analysis of different material models for polyurethane elastomer using estimation data sets


Sheikhi M. R., Shamsadinlo B., ÜNVER Ö., GÜRGEN S.

JOURNAL OF THE BRAZILIAN SOCIETY OF MECHANICAL SCIENCES AND ENGINEERING, cilt.43, sa.12, 2021 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 43 Sayı: 12
  • Basım Tarihi: 2021
  • Doi Numarası: 10.1007/s40430-021-03279-9
  • Dergi Adı: JOURNAL OF THE BRAZILIAN SOCIETY OF MECHANICAL SCIENCES AND ENGINEERING
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Aerospace Database, Communication Abstracts, Compendex, INSPEC, Metadex, Civil Engineering Abstracts
  • Anahtar Kelimeler: Finite element method, Data set estimation, Hyperelastic material, Polyurethane elastomer, STRAIN-ENERGY FUNCTION, CONSTITUTIVE MODEL, HYPERELASTIC MODELS, RUBBER, ELASTICITY, BEHAVIOR
  • Eskişehir Osmangazi Üniversitesi Adresli: Evet

Özet

When hyperelastic materials are included in a finite element analysis model, researchers generally have little adequate data to help them achieve their findings. Fortunately, data from tension or compression stress-strain testing are available for mostly researchers. This information must be analyzed and used to hyperelastic model studies. Curve fitting of these data is very crucial for determining the material constants. In the present study, first the experimental data sets were prepared and then used them to examine different hyperelastic material models by ANSYS. Experimental data of uni-axial tension, bi-axial tension, and pure shear tests are required to get an acceptable polynomial fit to the whole data set. Valanis-Landel (VL) method was utilized to estimate the shear and bi-axial data set by using the experimental uni-axial tension and uni-axial compression stress-strain data sets. The finite element analysis of specimens under uni-axial tension and pure shear tests were evaluated using four separate material models (Mooney-Rivlin, Ogden, neo-Hookean, and Gent). Each material model has a different form of strain energy density (SED) function. Constants of material models are extracted using a curve fitting tool utilizing the experimental data sets of uni-axial tension tests and the estimated data set of shear and bi-axial tests. These data sets were applied in different combinations (uni-axial tension + shear, uni-axial tension + bi-axial, shear + bi-axial, uni-axial tension + shear + bi-axial) to discover the material model constants. Finally, the best material model and combination form were selected. The optimal material model and combination form are picked at the end of the research. According to the results, Gent model rises as the best model for uni-axial stress testing, while neo-Hookean is the best choice for pure shear testing.