Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Method for calculating methanol molecular energy through deep learning

A deep learning and methanol technology, applied in chemical statistics, computational theoretical chemistry, chemical machine learning, etc., can solve problems such as difficult to obtain satisfactory results, and achieve the effect of avoiding convergence problems and reducing impact

Inactive Publication Date: 2019-02-12
DALIAN UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Restricted by this feature, it is difficult to obtain satisfactory results in the application of complex molecular systems or multi-molecular systems.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method for calculating methanol molecular energy through deep learning
  • Method for calculating methanol molecular energy through deep learning
  • Method for calculating methanol molecular energy through deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] A method to calculate the energy of methanol molecules through deep learning, using the spatial coordinates and corresponding energies of 1000 different configurations of methanol molecules. Preferably, the methyl portion in the methanol molecule of the present invention has undergone a fixed configuration treatment, that is, the spatial configuration of the carbon atom on the methyl group and the three hydrogen atoms (i.e. H1, H2, H3) remains unchanged, and the three H-C-H (ie H1-C-H2, H2-C-H3, H3-C-H1) bond angles are kept at 109 ° 28', three C-H bond lengths are For the carbon-oxygen bond in the methanol molecule, the direction of the bond axis is kept perpendicular to the plane formed by the three hydrogen atoms of the methyl group, while the spatial position of the oxygen atom changes along the direction of the carbon-oxygen bond axis to form different bond lengths. Variations range from For the hydroxyl bond in the methanol molecule, ie O-H4, the bond length va...

Embodiment 2

[0050] This embodiment provides a preferred solution of the hardware platform and software environment of the present invention.

[0051]Choose the low-end i5-6500CPU@3.20GHz / NVIDIA Corporation GK208[GeForce GT730] / 4G Mem hardware platform to obtain higher general performance; the software environment is Linux kernel 4.9 / TensorFlow-GPU 1.8.0 (installed via pip ), the driver is CUDA 9.0 / cuDNN 7.1.

Embodiment 3

[0053] This embodiment provides a preferred scheme for selecting input data in the present invention.

[0054] The constructed data set contains the configurations of 1000 methanol molecules and their one-to-one corresponding energies. The configurations are expressed in Bohrpositions, and the energy unit is kJ / mol. This method avoids using the gradient descent method to calculate the minimized total energy while ensuring that the energy accuracy of methanol molecules increases with the training set, and compares the results with those calculated using the standard DFT approximation (PBE). Methanol molecule parameters are set to 11, respectively: r C-O (C-O bond length), r O-H4 (O-H4 bond length), θ (C-O-H4 bond angle), cosφ (φ is the H1-C-O-H4 dihedral angle, cos is the trigonometric cosine function), the reciprocal of the distance between atoms, including: 1 / r C-O ,1 / r O-H4 ,1 / r C-H4 ,1 / r O-H1 ,1 / r H4-H1 ,1 / r H4-H2 ,1 / r H4-H3 . According to the PBE results, the opti...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a method for calculating methanol molecular energy through deep learning, and belongs to the technical field of molecular energy calculation. The method comprises the followingsteps: S1, building a methanol molecular database; S2, randomly selecting m configurations and corresponding energy as a training group, and taking the rest of 1000 to m configurations and corresponding energy as a test group; S3, transforming coordinates of a methanol molecular space in the training group into 11 parameters for building training group configuration parameter input matrixes; S4,extracting the energy data of the training group as a training group output energy matrixes corresponding to the training group configuration parameter input matrixes one by one; S5, building test group configuration parameter input matrixes and test group output energy matrixes, wherein the matrixes include 1000 to m lines; S6, learning the methanol molecular energy by adopting a dual-nervous layer calculation structure. By adopting the method, the influence of the ratio of the training group to the test group on the training result accuracy is lowered.

Description

technical field [0001] The invention relates to the technical field of molecular energy calculation, in particular to a method for calculating the energy of methanol molecules through deep learning. Background technique [0002] The machine learning of contemporary artificial intelligence, in the field of molecular structure optimization and minimum energy calculation, has a remarkable feature that as the degree of freedom of the initial matrix increases, that is, the more complex the molecular structure, the greater the degree of freedom, and the training group and the test group must be increased. ratio for accurate results. Restricted by this feature, it is difficult to obtain satisfactory results in the application of complex molecular systems or multi-molecular systems. Contents of the invention [0003] In order to solve the defects in the prior art, the present invention provides a method for calculating the energy of methanol molecules through deep learning, which...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G16C20/70G16C10/00
Inventor 周立川崔洪光王旋魏俊峰
Owner DALIAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com