Song Keke, Zhao Rui, Liu Jiahui, Wang Yanzhou, Lindgren Eric, Wang Yong, Chen Shunda, Xu Ke, Liang Ting, Ying Penghua, Xu Nan, Zhao Zhiqiang, Shi Jiuyang, Wang Junjie, Lyu Shuang, Zeng Zezhu, Liang Shirong, Dong Haikuan, Sun Ligang, Chen Yue, Zhang Zhuhua, Guo Wanlin, Qian Ping, Sun Jian, Erhart Paul, Ala-Nissila Tapio, Su Yanjing, Fan Zheyong
Beijing Advanced Innovation Center for Materials Genome Engineering, University of Science and Technology Beijing, Beijing, P. R. China.
School of Materials Science and Engineering, Hunan University, Changsha, China.
Nat Commun. 2024 Nov 25;15(1):10208. doi: 10.1038/s41467-024-54554-x.
Machine-learned potentials (MLPs) have exhibited remarkable accuracy, yet the lack of general-purpose MLPs for a broad spectrum of elements and their alloys limits their applicability. Here, we present a promising approach for constructing a unified general-purpose MLP for numerous elements, demonstrated through a model (UNEP-v1) for 16 elemental metals and their alloys. To achieve a complete representation of the chemical space, we show, via principal component analysis and diverse test datasets, that employing one-component and two-component systems suffices. Our unified UNEP-v1 model exhibits superior performance across various physical properties compared to a widely used embedded-atom method potential, while maintaining remarkable efficiency. We demonstrate our approach's effectiveness through reproducing experimentally observed chemical order and stable phases, and large-scale simulations of plasticity and primary radiation damage in MoTaVW alloys.
机器学习势（MLP）已展现出卓越的准确性，但缺乏适用于广泛元素及其合金的通用MLP限制了它们的适用性。在此，我们提出了一种有前景的方法，用于构建适用于众多元素的统一通用MLP，并通过一个针对16种元素金属及其合金的模型（UNEP-v1）进行了演示。为了实现化学空间的完整表示，我们通过主成分分析和多样的测试数据集表明，采用单组分和双组分体系就足够了。与广泛使用的嵌入原子法势相比，我们的统一UNEP-v1模型在各种物理性质方面表现出卓越的性能，同时保持了显著的效率。我们通过再现实验观察到的化学有序和稳定相，以及对MoTaVW合金的塑性和初级辐射损伤进行大规模模拟，证明了我们方法的有效性。
