Fig. 2 | npj Computational Materials

Fig. 2

From: Developing an interatomic potential for martensitic phase transformations in zirconium by machine learning

Fig. 2

Comparison of the potential energy predicted using the ML potential compared to AIMD calculations for a β-Zr supercell, b α-Zr supercell, c ω-Zr supercell, and d hypothetical fcc-Zr supercell. Note that the β-Zr, α-Zr, and ω-Zr data are included in the “training” dataset. A perfect correlation with the DFT values would correspond to the black lines. MAE represents mean absolute error

Back to article page