Modelling Polymers Under Impact Loading: A Fast Calibration Strategy Using Machine Learning
; Torres, J. P. ; Jerabek, M. ; Major, Z. Otros responsables
Fecha
06/2025
Editorial
Springer
Revista
Journal of Dynamic Behavior of Materials
Título del libro
Nombre del evento
Fecha del evento
Institución organizadora
ISSN
2199-7446
eISSN
2199-7454
ISBN
Idioma
Inglés
Tipo del evento
Clasificación temática
Palabras clave
Resumen
For polymers, inelastic deformation and failure under dynamic loading is largely dependent on temperature, strain rateand stress multiaxiality. Consequently, modelling their impact behavior requires several experiments and time-consumingdetermination of material parameters. In this work, a calibration strategy to reduce calibration times and experimental work-loads based on neural networks and genetic algorithms is presented. The method is demonstrated on Charpy- and punctureexperiments at 23 ◦ C and 50 ◦ C with two additional tensile tests at − 30 ◦ C and 23 ◦ C. In many cases, these experiments areconducted for datasheet specifications and are readily available. To achieve a balance between precision, ability to generalizeand numerical simplicity, the strain-rate and temperature dependence was mapped via a Johnson–Cook model that createstabular inputs for a general J2 plasticity model in Abaqus. Similarly, failure behavior was modelled using ductile-damagein Abaqus. The traditional fitting methodology based on inverse analysis using finite element simulations was replaced bythe use of neural network surrogate models which allow for calibration in a matter of minutes. This approach was validatedagainst four-point bending experiments with geometry, velocity and temperatures different to those used to calibrate themodel on. Here, simulated toughness and limit loads on average differed by 19% and 5% from experimental results.
Colecciones
Aprobación
Revisión
Complementado por
Referenciado por
Licencia Creative Commons
Excepto donde se indique lo contrario, la licencia de este ítem se describe como by-nc-sa
