卷积神经网络以协助评估X射线粉末衍射的格子参数
Juan Iván Gómez-Peralta1, Xim Bokhimi2, Patricia Quintana1
1Laboratorio Nacional de Nano y Biomateriales, CINVESTAV-IPN, Antigua Carretera a Progreso km 6, A. P. 37, 97310 Mérida, Yucatán, Mexico.
卷积神经网络 (CNN) 准确地估计有机晶体中的格子参数. 纳入原子组成数据显著提高了CNN准确度,用于从衍射模式预测单元细胞向量,角度和体积.
科学领域:
- 晶体学和材料科学 材料科学
- 计算化学和机器学习
背景情况:
- 准确确定格子参数对于理解和预测材料特性至关重要.
- 传统的格子参数估计方法可能是耗时和计算密集的.
研究的目的:
- 开发和评估卷积神经网络 (CNN) 用于在有机化合物中自动估计晶格参数.
- 研究将原子组成数据纳入CNN模型以提高准确性的影响.
主要方法:
- 在92,085个有机化合物的大型数据集上训练了两个CNN架构 (XRD-CNN和XRDElem-CNN).
- 为训练和测试生成模拟的X射线衍射 (XRD) 模式.
- 作为输入,XRDElem-CNN使用了衍射模式和单元细胞原子的二进制表示.
主要成果:
- XRD-CNN 实现了中等准确度 (MAPE:向量为 11.04%,角度为 7.40%,体积为 26.83%).
- XRDElem-CNN显示了显著提高的准确性 (MAPE:向量为4.73%,角度为6.49%,体积为6.05%).
- 使用真实XRD数据和Lp搜索方法验证了XRDElem-CNN性能.
结论:
- 采用原子信息的CNN,特别是XRDElem-CNN,为格子参数估计提供了强大而高效的方法.
- 这种方法对高通量材料的发现和表征有很大的前景.
- 拟议的原子表示对CNN评估具有计算效率.
更多相关视频
09:16X-ray Powder Diffraction in Conservation Science: Towards Routine Crystal Structure Determination of Corrosion Products on Heritage Art Objects
Published on: June 8, 2016
08:44Measurements of Long-range Electronic Correlations During Femtosecond Diffraction Experiments Performed on Nanocrystals of Buckminsterfullerene
Published on: August 22, 2017
相关概念视频
X-ray Crystallography
Diffraction
Diffraction is the change in the direction of travel experienced by an electromagnetic wave when it encounters a physical barrier whose dimensions are comparable to those of the wavelength of the light. X-rays are electromagnetic radiation with wavelengths about as long as the distance between neighboring...
X-ray Diffraction of Biological Samples
According to Bragg's law, when X-rays strike the sample positioned on a stage, the rays are scattered by the electron clouds around the sample atoms. The X-ray diffraction or scattering is caused by constructive interference of the X-ray waves that reflect off the internal...
Lattice Centering and Coordination Number
Types of Unit Cells
Imagine taking a large number of identical...
Trends in Lattice Energy: Ion Size and Charge
