arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.09674cond-mat.mtrl-sci
收藏

通过机器学习通用势函数加速高通量声子计算

Accelerating High-Throughput Phonon Calculations via Machine Learning Universal Potentials

Huiju Lee, Vinay I. Hegde, Chris Wolverton, Yi Xia

AI总结:

本研究提出基于多原子团簇展开(MACE)的机器学习通用势函数方法,大幅加速高通量简谐声子计算,在保证高精度的同时减少所需超胞数量,还构建了高质量DFT数据集供相关研究使用。

摘要:

声子在决定多种材料特性方面发挥着关键作用,但传统的声子计算方法计算量极大,限制了其广泛应用。在本研究中,我们提出了一种利用机器学习通用势函数加速高通量简谐声子计算的方法。我们基于多原子团簇展开(MACE,multi-atomic cluster expansion),在一个包含77种元素的2738种晶体结构的综合数据集上训练了一种最先进的机器学习原子间势,该数据集总计包含15670个超胞结构,均采用高保真密度泛函理论(DFT,density functional theory)计算得到。我们的方法显著减少了声子计算所需的超胞数量,同时在预测各类材料的简谐声子特性方面保持了高精度。训练后的模型在包含384种材料的留出子集上进行了声子计算验证,在全声子色散的振动频率上达到0.18 THz的平均绝对误差(MAE),在300K下的亥姆霍兹振动自由能上达到2.19 meV/原子,同时在材料动力学稳定性的分类准确率上达到86.2%。对126个体系的多晶型稳定性进行的热力学分析表明,该模型在300K和1000K下的结果与DFT结果吻合良好。此外,本研究构建的多样且广泛的高质量DFT数据集,为研究人员训练和改进其他机器学习原子间势模型提供了宝贵资源。

英文摘要:

Phonons play a critical role in determining various material properties, but conventional methods for phonon calculations are computationally intensive, limiting their broad applicability. In this study, we present an approach to accelerate high-throughput harmonic phonon calculations using machine learning universal potentials. We train a state-of-the-art machine learning interatomic potential, based on multi-atomic cluster expansion (MACE), on a comprehensive dataset of 2,738 crystal structures with 77 elements, totaling 15,670 supercell structures, computed using high-fidelity density functional theory (DFT) calculations. Our approach significantly reduces the number of required supercells for phonon calculations while maintaining high accuracy in predicting harmonic phonon properties across diverse materials. The trained model is validated against phonon calculations for a held-out subset of 384 materials, achieving a mean absolute error (MAE) of 0.18 THz for vibrational frequencies from full phonon dispersions, 2.19 meV/atom for Helmholtz vibrational free energies at 300K, as well as a classification accuracy of 86.2% for dynamical stability of materials. A thermodynamic analysis of polymorphic stability in 126 systems demonstrates good agreement with DFT results at 300 K and 1000 K. In addition, the diverse and extensive high-quality DFT dataset curated in this study serves as a valuable resource for researchers to train and improve other machine learning interatomic potential models.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑