Regensburg 2022 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
MM: Fachverband Metall- und Materialphysik
MM 10: Poster Session 1
MM 10.4: Poster
Montag, 5. September 2022, 18:00–20:00, P2
Interatomic Potential Fitting in pyiron — •Marvin Poul1, Niklas Leimeroth2, Alexander Knoll3, Marius Herbold3, and Joerg Neugebauer1 — 1Max-Planck-Institut fuer Eisenforschung — 2Universitaet Goettingen — 3TU Darmstadt
Interatomic potentials have a key role in computational materials science bridging the gap between Ab-Initio methods and large-scale engineering applications. pyiron[1] is an IDE for computational materials science that allows for reproducible yet easy to write simulation protocols. Within its framework we have developed tools that allow for semi-automatic fitting and verifying of potentials. This includes wrappers for AtomicRex[2] (for classical potentials) and as well the machine-learned Moment Tensor Potentials (MTP[3]), high-dimensional neural network potentials (HDNNP[4]), and ACE potentials. In this work we introduce the facilities offered by our code on the examples of an Magnesium MTP and an EAM potential for Copper.
[1]: https://doi.org/10.1016/j.commatsci.2018.07.043
[2]: https://doi.org/10.1088/1361-651X/aa6ecf
[3]: https://doi.org/10.1088/2632-2153/abc9fe
[4]: https://doi.org/10.1103/PhysRevLett.98.146401