arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2111.03563physics.chem-phcs.LG
收藏

机器学习预测反应性原子-双原子碰撞系统从初始反应物态到产物态分布

Machine Learning Product State Distributions from Initial Reactant States for a Reactive Atom-Diatom Collision System

University of Basel(巴塞尔大学);Indian Institute of Science Education and Research (IISER) Tirupati(印度科学教育与研究学院蒂鲁帕蒂分院);Stanford University(斯坦福大学);Air Force Research Laboratory(美国空军研究实验室)

Julian Arnold, Juan Carlos San Vicente Veliz, Debasish Koner, Narendra Singh, Raymond J. Bemish, Markus Meuwly

AI总结:

本研究提出机器学习STD模型,从初始反应物态预测产物态分布,经QCT数据训练,精度高且优于LB模型,适用于非平衡高速流模拟。

摘要:

提出了一种机器学习(ML)模型,用于预测反应性原子-双原子碰撞中从特定初始态到产物态分布(态到分布,即STD)的映射,并针对N($^4$S)+O$_{2}$(X$^3 Σ_{\rm g}^{-}$) $\ ightarrow$ NO(X$^2Π$) +O($^3$P)反应进行了定量测试。用于训练神经网络(NN)的参考数据集由显式准经典轨迹(QCT)模拟确定的最终态分布组成,涵盖约2000个初始条件。总体而言,以参考QCT与STD模型预测之间的均方根差异(约0.003)和$R^2$(约0.99)衡量的预测精度,在测试集、非网格特定初始态以及从具有平动、转动和振动温度特征的反应物态分布中抽取的初始条件下均较高。与在同一初始态分布上评估的更粗粒度的分布到分布(DTD)模型相比,STD模型表现出相当的性能,并额外具有反应物制备中态分辨的优势。从特定初始态出发还导致更多样化的最终态分布,这要求使用比DTD更具表达力的神经网络。显式QCT模拟、STD模型和广泛使用的Larsen-Borgnakke(LB)模型之间的直接比较表明,STD模型是定量的,而LB模型对于转动分布$P(j')$至多是定性的,并且对于振动分布$P(v')$则失败。因此,STD模型非常适合模拟非平衡高速流动,例如使用直接模拟蒙特卡洛方法。

英文摘要:

A machine learned (ML) model for predicting product state distributions from specific initial states (state-to-distribution or STD) for reactive atom-diatom collisions is presented and quantitatively tested for the N($^4$S)+O$_{2}$(X$^3 Σ_{\rm g}^{-}$) $\rightarrow$ NO(X$^2Π$) +O($^3$P) reaction. The reference data set for training the neural network (NN) consists of final state distributions determined from explicit quasi-classical trajectory (QCT) simulations for $\sim 2000$ initial conditions. Overall, the prediction accuracy as quantified by the root-mean-squared difference $(\sim 0.003)$ and the $R^2$ $(\sim 0.99)$ between the reference QCT and predictions of the STD model is high for the test set and off-grid state specific initial conditions and for initial conditions drawn from reactant state distributions characterized by translational, rotational and vibrational temperatures. Compared with a more coarse grained distribution-to-distribution (DTD) model evaluated on the same initial state distributions, the STD model shows comparable performance with the additional benefit of the state resolution in the reactant preparation. Starting from specific initial states also leads to a more diverse range of final state distributions which requires a more expressive neural network to be used compared with DTD. Direct comparison between explicit QCT simulations, the STD model, and the widely used Larsen-Borgnakke (LB) model shows that the STD model is quantitative whereas the LB model is qualitative at best for rotational distributions $P(j')$ and fails for vibrational distributions $P(v')$. As such the STD model can be well-suited for simulating nonequilibrium high-speed flows, e.g., using the direct simulation Monte Carlo method.

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。

↑