Automatic parameter identification of a shape memory alloy model using characteristic experimental data points

Research output: Contribution to journalResearch articleContributedpeer-review



Identifying material parameters of constitutive models can be a time consuming process. This is especially evident when the constitutive models have many parameters. In this work, an automatic identification procedure is described to determine the various parameters of a shape memory alloy model. The procedure requires some parameters as manual inputs (to be determined via trial and error) and determines some of the parameters directly from experimental data. In addition, the procedure calculates some of the parameters using Newton's method based on characteristic data points so that the simulation curves pass through those points. As a result, the method essentially decouples the effects of the parameters on the simulation uniaxial test curves. The automatic identification procedure is applied for uniaxial tests dealing with transformation, tension–compression asymmetry, internal loops, plasticity and functional fatigue. The procedure is fast, flexible and reasonably accurate. However, the approach is currently limited to uniaxial tests, small strains and needs to be tailored to each model individually.


Original languageEnglish
Article number105300
Number of pages12
JournalEuropean Journal of Mechanics. A, Solids
Volume106 (2024)
Publication statusPublished - 16 Mar 2024

External IDs

Scopus 85187958511
Mendeley e1b657a2-40fa-3813-a93f-3c03372c7193



  • Constitutive model, Functional fatigue, Material parameter identification, Plasticity, Shape memory alloys

Library keywords