Experimental validation of shape memory material model implemented in commercial finite element software under multiaxial loading

Abstract

Shape memory alloys are used in ever-increasing numbers of applications, such as implants made of porous shape memory alloys, where the material is subjected to complex loading conditions with various loading paths. Finite element simulation of such parts requires utilizing a constitutive model that is able to capture the multiaxial and path-dependent behavior of shape memory alloys. The main objective of this article is to investigate the accuracy of the constitutive model implemented in current commercial finite element software such as Ansys in predicting the shape memory alloys mechanical response under different multiaxial loading paths. To this end, several isothermal tests were conducted on thin-walled NiTi tubes with uniaxial, as well as multiaxial, path-varying loadings. The performance of the material model within Ansys was then investigated by finite element modeling of the sample tubes and performing simulations of the tests. Comparing the finite element results with experimental data, it was observed that while this model provided a close prediction of the uniaxial tensile superelastic response, it was not able to reproduce the multiaxial and path-dependent behavior of the shape memory alloy samples with sufficient accuracy. A brief discussion of the reasons behind the inaccuracy of the current model and potentially promising models for future investigation are provided.