New methods for prediction of elastic constants based on density functional theory combined with machine learning
New methods for prediction of elastic constants based on density functional theory combined with machine learning
复制标题
基于密度泛函理论结合机器学习的弹性常数预测新方法
DOI：
10.1016/j.commatsci.2017.06.015
复制
发表时间：
2017-10
影响因子：
3.3
通讯作者：
Zongguo Wang
中科院分区：
文献类型：
--
作者：
Juan Wang;Xiao-yu Yang;Zhi Zeng;Xiao-li Zhang;Xu-shan Zhao;Zongguo Wang
弹性常数在力学性能研究中起着重要作用，但其测量往往比较困难。虽然密度泛函理论（DFT）计算提供了一个可靠的方法来满足这一挑战，结果包含由各种近似引起的固有误差。机器学习的数据驱动方法也为预测材料性能奠定了基础。为了提高理论计算结果的准确性，本文研究了使用机器学习方法来校正DFT计算的弹性常数，并直接预测弹性常数。采用反向传播算法训练的单隐层前向神经网络（SLFN）、广义回归神经网络（GRNN）和支持向量机回归（SVR）技术，建立回归模型，对金属或金属二元合金的DFT弹性常数进行修正。我们还建立了回归模型来预测立方晶系金属二元合金的弹性常数，而不是使用DFT计算。结果表明，回归模型修正后的弹性常数比DFT计算的弹性常数有更高的精度，采用SLFN技术直接预测二元合金的弹性常数是有前景的。
Elastic constants play critical roles in researching mechanical properties, but they are usually difficult to be measured. While density functional theory (DFT) calculations provide a reliable method to meet this challenge, the results contain inherent errors caused by various approximations. The data-driven approach of machine learning also laid a foundation for predicting material properties. In order to increase the accuracy of theoretical calculations results, in this paper we investigate using machine learning methods to both correct the elastic constants by DFT calculation, and to directly predict elastic constants. The single-hidden layer feedforward neural network trained by back propagation algorithm (SLFN), general regression neural network (GRNN) and support vector machine for regression (SVR) techniques are employed to build regression models to correct the elastic constants by DFT calculation for metal or metallic binary alloys. We also build regression models to predict the elastic constants of metallic binary alloys with cubic crystal system rather than using DFT calculations. It has been demonstrated that the elastic constants corrected by regression models has higher accuracy than those calculated by DFT, and the elastic constants of binary alloys directly predicted by model using the outperformed SLFN technique is prospective.
登录
查看更多内容
DOI：
10.1016/0001-6160(61)90175-4
发表时间：
1961-12
期刊：
Acta Metallurgica
影响因子：
--
作者：
J. Trivisonno;Charles S. Smith
通讯作者：
J. Trivisonno;Charles S. Smith
影响因子：
3.2
作者：
Y. Chang;L. Himmel
通讯作者：
Y. Chang;L. Himmel
DOI：
10.7146/dpb.v19i339.6570
发表时间：
1990-11
期刊：
--
影响因子：
--
作者：
M. F. Møller
通讯作者：
M. F. Møller
影响因子：
--
作者：
Zhuo Wang;Xiaoyu Yang;Yufei Zheng;Q. Yong;Hang Su;Caifu Yang
通讯作者：
Zhuo Wang;Xiaoyu Yang;Yufei Zheng;Q. Yong;Hang Su;Caifu Yang
影响因子：
64.8
作者：
RUMELHART, DE;HINTON, GE;WILLIAMS, RJ
通讯作者：
WILLIAMS, RJ