Method for predicting whether solid solution is formed or not based on atom combination data

A technology for data prediction and solid solution, applied in the direction of chemical property prediction, etc., can solve the problems that thermodynamic methods cannot predict the formation of unknown solid solutions, the problems are complex, and the liquidus cannot be predicted, so as to avoid blindness, low cost, and easy implementation. Effect

Inactive Publication Date: 2021-11-12
上海帆阳信息科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the complexity of the problem, the thermodynamic method cannot predict whether the unknown solid solutio...

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 predicting whether solid solution is formed or not based on atom combination data
  • Method for predicting whether solid solution is formed or not based on atom combination data
  • Method for predicting whether solid solution is formed or not based on atom combination data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] In this example, the modeling results of whether a solid solution model is established based on the atomic combination data of 39 oxo-acid salt binary phase diagram samples combined with the support vector machine classification algorithm, are shown in Table 4.

[0036] Table 4 Modeling accuracy

[0037]

[0038] It can be seen from Table 4 that in the modeling process, among the 26 samples that can form solid solutions, 23 are accurately discriminated, with an accuracy rate of 88.46%. Among the 13 samples that did not form a solid solution, 12 were discriminated correctly, with an accuracy rate of 92.31%. The overall accuracy is 89.76%. It shows that the accuracy of the solid solution model established based on the support vector machine classification algorithm is relatively high.

Embodiment 2

[0040] For the 39 data, the internal cross-validation results of the leave-one-out method established based on the support vector machine classification algorithm for the formation of solid solution models are shown in Table 5. The leave-one-out method of internal cross-validation means that only one sample is left as the test set at a time, and the other samples are used as the training set. There are a total of 39 samples, which requires 39 training and 39 testing times.

[0041] Table 5 Leave-one-out accuracy

[0042]

[0043] It can be seen from Table 5 that in the leave-one-out internal cross-validation process, among the 26 samples that can form solid solutions, 21 are accurately discriminated, with an accuracy rate of 80.77%. Among the 13 samples that did not form a solid solution, 11 were accurately discriminated, with an accuracy rate of 84.62%, and the overall accuracy rate was 82.51%. It shows that the accuracy of leave-one-out method based on the support vector...

Embodiment 3

[0045] In this example, the established solid solution model is used to predict the three newly collected samples, and the results are shown in Table 6.

[0046] Table 6 Forecast results

[0047]

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

Prediction of macroscopic properties of substances is fundamentally a quantum mechanics and statistical mechanics solving problem. However, the problem is too complex, and the thermodynamic method cannot forecast whether an unknown solid solution is formed. According to the method, atomic data is tried to be combined, and then a data mining technology is utilized to establish whether a phase diagram forms a solid solution or not based on atomic combination data, especially whether an oxysalt binary phase diagram forms a solid solution or not. According to the atom combination data-data mining method, the tedious intermediate step of solving a thermodynamic function through atom parameters is omitted, an atom parameter set is directly mapped to a certain phase diagram feature, a simple and clear semi-empirical function is obtained, and a series of phase diagram laws are well summarized.

Description

technical field [0001] The invention relates to the field of phase diagram calculation of binary phase diagrams, in particular to a method for predicting whether a solid solution is formed based on atomic combination data. Background technique [0002] The phase diagram is a reflection of the thermodynamic properties of the phase, and the thermodynamic properties are the statistical results of the interaction between particles and the resulting motion and dynamic structure. Many molten salt phase diagrams have extensive solid solution formation regions. The formation of solid solutions has a great influence on the liquidus (liquid surface) and the shape of the phase diagram. Some oxo-acid salt molten salt solutions are widely used in high-temperature fuel cells, heat transfer agents, heat treatment media, crystal growth media, etc. Some solid solutions formed by oxo-acid salt systems are useful optical, electrical, and magnetic materials. Many rock minerals also exist in t...

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
IPC IPC(8): G16C20/30
CPCG16C20/30
Inventor 刘振昌刘太行刘太昂吴治富周央周晶晶朱峰刘远刘婷婷朱鲁阳
Owner 上海帆阳信息科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap