An inverse finite element approach for material characterization of small hyperelastic samples.

Viljoen, Marga Christine (2019-12)

Thesis (MEng)--Stellenbosch University, 2019.

Thesis

ENGLISH ABSTRACT: The aim of this research was to develop and test an experimental method, which could be scaled to samples of 4 mm x 4 mm x 1-2 mm in size. To validate the design of the experimental method, it was applied to samples (10 mm by 10 mm by 2.5 mm in size) of a soft hyperelastic silicone-rubber using available equipment. The biggest challenge in this research was the small scale at which testing was performed. In addition, the size of force measurements was small (under 10 N), which meant that force measurements inherently included noise, which could not be neglected. The method was designed such that stress concentrations were avoided and boundary conditions were simple enough to be easily implemented in a finite element analysis. A uni-axial compression test was performed where full-field displacement data was measured using digital image correlation and force data was measured using a 50 N load cell. Displacement and force results for compressions of between 20-55 % of the height of the sample were used in an inverse finite element model updating method to iteratively determine a three parameter Mooney-Rivlin material model for the hyperelastic silicone-rubber. There was a large amount of variability in the force results, even though the displacement inputs for the tests were repeatable. The inverse finite element model updating method showed repeatability in the value to which it converged, however the material model obtained from the optimization process produced accurate displacements neither when plotted against the experimental displacements nor when implemented in a validation test geometry.

AFRIKAANSE OPSOMMING: Die doel van hierdie navorsing was om ’n eksperimentele metode, wat geskaal kan word na monsters van 4 mm x 4 mm 1-2 mm, te ontwikkel en te toets. Om die ontwerp te toets, is die metode toegepas op monsters (10 mm by 10 mm by 2,5 mm in grootte) gemaak van ’n sagte hiper-elastiese silikoon-rubber deur gebruik te maak van beskikbare toerusting. Die grootste uitdaging in hierdie navorsing was die klein skaal waarby toetse uitgevoer is. As gevolg daarvan, was die grootte van gemete kraglesings klein (onder 10 N), wat beteken het dat resultate geraas ingesluit het, wat nie weglaatbaar klein was nie. Die metode is op so manier ontwerp dat spanningskonsentrasies vermy is en randvoorwaardes eenvoudig genoeg was om maklik in ’n eindige element analise geïmplementeer te word. ’n Eenassige druktoets is uitgevoer waar volveldverplasingdata gemeet is deur digitale beeldkorrelasie en kraglesings geneem is met behulp van ’n 50 N lassel. Verplasings en kragte vir vervormings van tussen 20-55 % van die hoogte van die monster is gebruik in ’n inverse eindige element model opdateringsmetode om die konstantes van ’n drie-parameter Mooney- Rivlin materiaal model vir die hiper-elastiese silikoon-rubber te bepaal. Daar was ’n groot mate van variasie in die kraglesings, alhoewel die verplasingsinsette herhaalbaar was. Die inverse eindige element model opdateringsmetode het herhaalbaar tot dieselfde waarde konvergeer, maar die materiaalmodel wat uit die optimeringsproses verkry is, het onakkurate verplasingswaardes voorspel vir beidie die experimentele geometrie en die validasie toets.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/107307
This item appears in the following collections: