{"title":"利用神经网络波函数计算固体的力和应力","authors":"Yubing Qian, Xiang Li, Ji Chen","doi":"10.1039/d4fd00071d","DOIUrl":null,"url":null,"abstract":"Accurate <em>ab initio</em> calculations of real solids are of fundamental importance in fields such as chemistry, phases and materials science. Recently, variational Monte Carlo (VMC) based on neural network wavefunction has been developed as a promising option to solve the existing challenges in <em>ab initio</em> calculations. In this study, we discuss the calculation of interatomic force and stress tensor of real solids with neural network--based VMC method. A new scheme of computing force is proposed based on the space warp coordination transformation method, which achieves better accuracy, efficiency and robustness than existing methods. In addition, we also designed new periodic features of neural network to further improve the robustness of force calculations for different lattices. This work paves the way for further extending the application of machine learning quantum Monte Carlo in materials modelling.","PeriodicalId":76,"journal":{"name":"Faraday Discussions","volume":null,"pages":null},"PeriodicalIF":3.3000,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Force and stress calculation with neural network wavefunction for solids\",\"authors\":\"Yubing Qian, Xiang Li, Ji Chen\",\"doi\":\"10.1039/d4fd00071d\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Accurate <em>ab initio</em> calculations of real solids are of fundamental importance in fields such as chemistry, phases and materials science. Recently, variational Monte Carlo (VMC) based on neural network wavefunction has been developed as a promising option to solve the existing challenges in <em>ab initio</em> calculations. In this study, we discuss the calculation of interatomic force and stress tensor of real solids with neural network--based VMC method. A new scheme of computing force is proposed based on the space warp coordination transformation method, which achieves better accuracy, efficiency and robustness than existing methods. In addition, we also designed new periodic features of neural network to further improve the robustness of force calculations for different lattices. This work paves the way for further extending the application of machine learning quantum Monte Carlo in materials modelling.\",\"PeriodicalId\":76,\"journal\":{\"name\":\"Faraday Discussions\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":3.3000,\"publicationDate\":\"2024-04-10\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Faraday Discussions\",\"FirstCategoryId\":\"92\",\"ListUrlMain\":\"https://doi.org/10.1039/d4fd00071d\",\"RegionNum\":3,\"RegionCategory\":\"化学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"CHEMISTRY, PHYSICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Faraday Discussions","FirstCategoryId":"92","ListUrlMain":"https://doi.org/10.1039/d4fd00071d","RegionNum":3,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"CHEMISTRY, PHYSICAL","Score":null,"Total":0}
引用次数: 0
摘要
在化学、相学和材料科学等领域,对实际固体进行精确的 ab initio 计算至关重要。最近,基于神经网络波函数的变分蒙特卡罗(VMC)作为一种有前途的选择被开发出来,以解决现有的非初始计算难题。在本研究中,我们讨论了用基于神经网络的 VMC 方法计算实际固体的原子间力和应力张量。我们提出了一种基于空间翘曲协调变换方法的力计算新方案,与现有方法相比,该方案具有更高的精度、效率和鲁棒性。此外,我们还设计了新的神经网络周期特征,进一步提高了不同晶格下力计算的鲁棒性。这项工作为进一步扩展机器学习量子蒙特卡洛在材料建模中的应用铺平了道路。
Force and stress calculation with neural network wavefunction for solids
Accurate ab initio calculations of real solids are of fundamental importance in fields such as chemistry, phases and materials science. Recently, variational Monte Carlo (VMC) based on neural network wavefunction has been developed as a promising option to solve the existing challenges in ab initio calculations. In this study, we discuss the calculation of interatomic force and stress tensor of real solids with neural network--based VMC method. A new scheme of computing force is proposed based on the space warp coordination transformation method, which achieves better accuracy, efficiency and robustness than existing methods. In addition, we also designed new periodic features of neural network to further improve the robustness of force calculations for different lattices. This work paves the way for further extending the application of machine learning quantum Monte Carlo in materials modelling.