Density prediction method based on molecular dynamics and ridge regression algorithm

A technology of molecular dynamics and ridge regression, applied in design optimization/simulation, instrumentation, electrical digital data processing, etc., can solve the problems of no specific proposal and insufficient accuracy, and achieve high accuracy, accurate glass density value, The effect of simple method

Active Publication Date: 2021-08-27
NANJING FIBERGLASS RES & DESIGN INST CO LTD
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although the literature uses molecular dynamics, it only simulates the performance of glass materials, and the accuracy is not high enough, and it does not specifically propose an intuitive and rapid density prediction method.

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
  • Density prediction method based on molecular dynamics and ridge regression algorithm
  • Density prediction method based on molecular dynamics and ridge regression algorithm
  • Density prediction method based on molecular dynamics and ridge regression algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0063] This embodiment provides a density prediction method based on molecular dynamics and ridge regression algorithm, see figure 1 , the method includes:

[0064] Step 1, collect density data of glass materials with different components, and construct a density database, which includes one-to-one mapped glass components and their corresponding densities;

[0065] Step 2, construct the descriptor containing "material gene" of oxide glass material density based on molecular dynamics, see figure 2 ;

[0066] 2-1 Construct atomic structure models of oxide crystals with different symmetries as unit cells for molecular dynamics calculations;

[0067] 2-2 For each unit cell structure constructed in step 2-1, perform molecular dynamics calculations to obtain the optimized unit cell energy E crystal ;

[0068] 2-3 Aiming at the structure of crystals with various symmetries optimized by molecular dynamics calculations; then obtain their performance parameter sets through further ...

Embodiment 2

[0081] The present embodiment provides a density prediction method based on molecular dynamics and ridge regression algorithm, the method comprising:

[0082] Step 1, collect density data of glass materials with different components, and construct a density database, which includes one-to-one mapped glass components and their corresponding densities;

[0083] For the collected density data, in practical applications, it also includes preprocessing, and the preprocessing includes:

[0084] For any pair of glass components, judge whether the following two conditions are true at the same time:

[0085] Condition 1: In the glass composition, the molar ratio difference of each oxide component is less than or equal to the first preset threshold, and the unit is percentage;

[0086] Condition 2: the difference between the densities of the two glasses is greater than a second preset threshold, and the unit is a percentage;

[0087] If it is established at the same time, the correspo...

Embodiment 3

[0172] This embodiment provides a density prediction system based on molecular dynamics and ridge regression algorithm, said system comprising:

[0173] The density database building module is used to collect density data of glass materials with different components and construct a density database, which includes one-to-one mapped glass components and their corresponding densities;

[0174] Descriptor building blocks for constructing descriptors containing "material genes" for oxide glass material densities based on molecular dynamics;

[0175] The training data building block is used to construct a training set, a verification set and a test set based on the density database obtained by building the density database building block and the descriptor constructed by the descriptor building block;

[0176] The model builds a training module, which is used to build a density prediction model based on the ridge regression model tree, and trains the built density prediction model ...

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 density prediction method based on molecular dynamics and a ridge regression algorithm, and belongs to the technical field of glass performance prediction through machine learning. According to the method, when a descriptor for machine learning is constructed, atomic structure models of oxide crystals with different symmetry are constructed based on molecular dynamics to serve as unit cells for molecular dynamics calculation, and the binding energy of each unit cation i and the bond length L of the cation i and nearest neighbor oxygen ions in the unit cells are selected as prediction performance parameters; meanwhile, the density prediction model is constructed based on the ridge regression model, rapid and accurate prediction of the glass density is achieved, and the research and development cost is saved to a great extent especially for research and development of some glass products with high density requirements.

Description

technical field [0001] The invention relates to a density prediction method based on molecular dynamics and ridge regression algorithm, and belongs to the technical field of machine learning prediction glass performance. Background technique [0002] With the gradual popularization of wireless charging and 5G, the exterior parts of mobile phones need to use non-metallic materials, and ceramic glass has super flexural strength, extraordinary fracture toughness, good rigidity, high wear resistance, signal non-shielding, etc. It is very suitable for the back cover of smart phones, and is an engineering material that conforms to the general trend of future appearance parts development. [0003] Density is an important performance parameter of engineering materials. From a macroscopic point of view, density is a measure of the ability of an object to resist elastic deformation. From a microscopic point of view, it is a reflection of the bonding strength between atoms, ions or mol...

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 & AuthorityApplications(China)
IPC IPC(8): G06F30/27G06F119/14
CPCG06F30/27G06F2119/14
Inventor赵明赵谦刘鑫陈阳匡宁
OwnerNANJING FIBERGLASS RES & DESIGN INST CO LTD