CN113421621A - Generation type countermeasure network driven silicon germanium super crystal lattice luminescent new material development technology - Google Patents
Generation type countermeasure network driven silicon germanium super crystal lattice luminescent new material development technology Download PDFInfo
- Publication number
- CN113421621A CN113421621A CN202110668704.6A CN202110668704A CN113421621A CN 113421621 A CN113421621 A CN 113421621A CN 202110668704 A CN202110668704 A CN 202110668704A CN 113421621 A CN113421621 A CN 113421621A
- Authority
- CN
- China
- Prior art keywords
- silicon
- superlattice
- germanium
- model
- generative
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 229910000577 Silicon-germanium Inorganic materials 0.000 title claims abstract description 28
- LEVVHYCKPQWKOP-UHFFFAOYSA-N [Si].[Ge] Chemical compound [Si].[Ge] LEVVHYCKPQWKOP-UHFFFAOYSA-N 0.000 title claims abstract description 26
- 239000000463 material Substances 0.000 title claims abstract description 22
- 238000005516 engineering process Methods 0.000 title claims abstract description 18
- 238000011161 development Methods 0.000 title claims abstract description 11
- 239000013078 crystal Substances 0.000 title abstract description 4
- 238000004364 calculation method Methods 0.000 claims abstract description 15
- 238000000034 method Methods 0.000 claims abstract description 15
- 238000012549 training Methods 0.000 claims abstract description 12
- 238000004088 simulation Methods 0.000 claims abstract description 10
- 238000013507 mapping Methods 0.000 claims abstract description 4
- 238000013528 artificial neural network Methods 0.000 claims description 2
- 238000013527 convolutional neural network Methods 0.000 claims description 2
- 230000000306 recurrent effect Effects 0.000 claims description 2
- 238000013461 design Methods 0.000 claims 1
- 230000004927 fusion Effects 0.000 claims 1
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 4
- 229910052710 silicon Inorganic materials 0.000 description 4
- 239000010703 silicon Substances 0.000 description 4
- 238000004020 luminiscence type Methods 0.000 description 3
- 238000004377 microelectronic Methods 0.000 description 2
- 239000000758 substrate Substances 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000005693 optoelectronics Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000000547 structure data Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C60/00—Computational materials science, i.e. ICT specially adapted for investigating the physical or chemical properties of materials or phenomena associated with their design, synthesis, processing, characterisation or utilisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/26—Composites
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
- Luminescent Compositions (AREA)
Abstract
The invention discloses a generation type countermeasure network driven silicon germanium super crystal lattice luminescent new material development technology, the method comprises: calculating and generating the data of the energy band structures of the silicon-germanium super-lattices in different arrangement modes by using a first principle; constructing a generative countermeasure network, and modeling a mapping relation of the superlattice corresponding to the energy band structure; constructing a training data set by using a sample generated by simulation calculation and a small amount of experimental data; performing staggered joint training on a generation model and a discrimination model in the generative confrontation network; searching silicon-germanium superlattice material systems in different arrangement modes by using a generation model, and selecting a structure with the optimal light-emitting performance; the generative model obtained by the method can analyze the energy band structure of the silicon-germanium superlattice at a speed far higher than that calculated by a first principle; the method can realize the search of a large-scale material structure and guide the development of a novel silicon-germanium luminescent material.
Description
Technical Field
The invention relates to the field of material calculation, in particular to a technology for developing a new generation type confrontation network-driven silicon-germanium superlattice luminescent material.
Background
The moore law based on the microelectronic technology is in failure in the near future, and the silicon-based optoelectronic on-chip integrated technology compatible with the current microelectronic CMOS technology is expected to become a cornerstone of the future information technology and continues the moore law. The light source on the silicon substrate lacking high-efficiency luminescence becomes the last obstacle of the integration technology on the silicon substrate photoelectron chip, and the successful development of the light source can lead the great revolution of the whole semiconductor chip technology. The development of high-efficiency silicon-germanium superlattice luminescent materials is an important breakthrough direction of the current silicon-based luminescent materials.
At present, the silicon germanium superlattice is theoretically designed and verified through first principle calculation, and experiments and tests are guided. This conventional method of operation faces two major problems. Firstly, the first principle has high calculation resource cost, and each calculation of a sample requires a large amount of time and calculation power, so that the candidate space capable of being searched is limited; the possible arrangement and combination modes of the silicon-germanium superlattice are very large, and effective results are difficult to obtain through limited search. Secondly, the calculation of the first principle is approximate calculation, the result has deviation from the actual condition, and the accuracy cannot be completely ensured.
Disclosure of Invention
The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a technology for developing a new generative countermeasure network-driven sige superlattice light-emitting material, which only needs to learn limited simulation samples and a small number of real samples to obtain the capability of evaluating key information. In the using stage, the judgment on the light emitting performance of the silicon germanium superlattice in different arrangement modes can be realized at a speed far higher than the calculation speed of the first principle. The technology fuses the information contained in the simulation sample and the real sample in the form of a generative confrontation network, and improves the accuracy on the basis of the calculation of a first principle. The technology is expected to find a new silicon-germanium superlattice material with excellent luminescence property in large-scale search.
The invention provides a generation type countermeasure network driven silicon germanium super crystal lattice luminescent new material development technology, the technical method comprises:
and (3) generating data of the energy band structures of the silicon-germanium superlattices in different arrangement modes by using a first principle calculation.
And constructing a generative countermeasure network, and modeling the mapping relation of the superlattice corresponding to the energy band structure.
Preferably, the generative model in the generative confrontation network is a recurrent neural network.
Preferably, the discriminant model in the generative countermeasure network is a convolutional neural network.
A training data set is constructed using samples generated by simulation calculations and a small amount of experimental data.
Furthermore, when a training data set is constructed, a simulation sample is taken as a main part, and a small amount of real samples are combined. The weights of both in training are inversely proportional to their numbers.
And performing staggered joint training on the generative model and the discriminant model in the generative confrontation network.
And searching silicon-germanium superlattice material systems with different arrangement modes by using the generation model, and selecting a structure with the optimal light-emitting performance.
Furthermore, in the searching process, the model can be continuously trained and upgraded by combining with the experimental test result, and the result accuracy is improved.
Preferably, the search process may be performed in parallel on multiple computing devices.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flow chart of the development technique of the new generative countermeasure network-driven SiGe superlattice light-emitting material of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention provides a new generation type countermeasure network driven silicon-germanium superlattice luminescent material development technology, which only needs to learn limited simulation samples and a small amount of real samples to obtain the judgment capability of carrying key information. In the using stage, the judgment on the light emitting performance of the silicon germanium superlattice in different arrangement modes can be realized at a speed far higher than the calculation speed of the first principle. The technology is expected to find a new silicon-germanium superlattice material with excellent luminescence property in large-scale search.
FIG. 1 is a flow chart of the development technique of the new generative countermeasure network-driven SiGe superlattice light-emitting material of the present invention.
As shown in fig. 1, the technology for developing new generative anti-network-driven silicon-germanium superlattice light-emitting material comprises the following steps:
and step 101, calculating and generating the energy band structure data of the silicon-germanium superlattice in different arrangement modes by using a first principle.
And 102, constructing a generative countermeasure network, and modeling the mapping relation of the superlattice corresponding to the energy band structure.
And 103, constructing a training data set by using the sample generated by the simulation calculation and a small amount of experimental data.
And 104, performing staggered joint training on the generative model and the discriminant model in the generative confrontation network.
And 105, searching silicon-germanium superlattice material systems with different arrangement modes by using the generated model.
It should be noted that, in the present specification, the embodiments are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (6)
1. A generation-type countermeasure network-driven silicon-germanium superlattice light-emitting new material development technology is characterized in that the method comprises the following steps:
calculating and generating the data of the energy band structures of the silicon-germanium super-lattices in different arrangement modes by using a first principle;
constructing a generative countermeasure network, and modeling a mapping relation of the superlattice corresponding to the energy band structure;
constructing a training data set by using a sample generated by simulation calculation and a small amount of experimental data;
performing staggered joint training on a generation model and a discrimination model in the generative confrontation network;
and searching silicon-germanium superlattice material systems with different arrangement modes by using the generation model, and selecting a structure with the optimal light-emitting performance.
2. The method of claim 1, wherein the energy band structure of the silicon germanium superlattice is calculated using first principles, and the arrangement of the superlattice is randomly generated without special manual design.
3. The method of claim 2, wherein the computing of the randomly arranged structure using the principle of primeness generates the samples in a manner that multiple machines execute in parallel.
4. The method of claim 1, wherein the generative model is a recurrent neural network and the discriminative model is a convolutional neural network.
5. The method according to claim 1, wherein the used training samples are combined with experimental data in the simulation data, and the training process can realize the fusion of information in the simulation samples and the real samples, so that the practicability of the model prediction result is improved.
6. The method of claim 1, wherein the random superlattice structure is searched using a generative model, and the output is a score indicating the probability that the energy band is a direct bandgap.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110668704.6A CN113421621A (en) | 2021-06-16 | 2021-06-16 | Generation type countermeasure network driven silicon germanium super crystal lattice luminescent new material development technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110668704.6A CN113421621A (en) | 2021-06-16 | 2021-06-16 | Generation type countermeasure network driven silicon germanium super crystal lattice luminescent new material development technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113421621A true CN113421621A (en) | 2021-09-21 |
Family
ID=77788757
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110668704.6A Pending CN113421621A (en) | 2021-06-16 | 2021-06-16 | Generation type countermeasure network driven silicon germanium super crystal lattice luminescent new material development technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113421621A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107574164A (en) * | 2017-09-12 | 2018-01-12 | 中国农业大学 | A kind of method for extracting nucleic acid based on coaxial capillary |
| CN109871610A (en) * | 2019-02-18 | 2019-06-11 | 中国科学院理化技术研究所 | Novel non-linearity optical material virtual screening system based on first principle |
| US20200333188A1 (en) * | 2019-04-16 | 2020-10-22 | Huazhong University Of Science And Technology | Material optical transition analysis method and system |
| CN111816266A (en) * | 2020-07-10 | 2020-10-23 | 北京迈高材云科技有限公司 | Method and system for automatically constructing material quantitative structural property model |
-
2021
- 2021-06-16 CN CN202110668704.6A patent/CN113421621A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107574164A (en) * | 2017-09-12 | 2018-01-12 | 中国农业大学 | A kind of method for extracting nucleic acid based on coaxial capillary |
| CN109871610A (en) * | 2019-02-18 | 2019-06-11 | 中国科学院理化技术研究所 | Novel non-linearity optical material virtual screening system based on first principle |
| US20200333188A1 (en) * | 2019-04-16 | 2020-10-22 | Huazhong University Of Science And Technology | Material optical transition analysis method and system |
| CN111816266A (en) * | 2020-07-10 | 2020-10-23 | 北京迈高材云科技有限公司 | Method and system for automatically constructing material quantitative structural property model |
Non-Patent Citations (1)
| Title |
|---|
| 刘刚、林建涵等: "激光控制平地系统设计与试验分析", 《农业机械学报》, vol. 37, no. 1, pages 71 - 74 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jyu et al. | Statistical timing analysis of combinational logic circuits | |
| Venkata Kamesh et al. | An innovative approach to detect isomorphism in planar and geared kinematic chains using graph theory | |
| Ward et al. | PADE: a high-performance placer with automatic datapath extraction and evaluation through high dimensional data learning | |
| US20090271761A1 (en) | Generative Programming System and Method Employing Focused Grammars | |
| CN105447031A (en) | Training sample labeling method and device | |
| Ponce-Cruz et al. | A practical approach to metaheuristics using LabVIEW and MATLAB® | |
| Rehbach et al. | Comparison of parallel surrogate-assisted optimization approaches | |
| CN105022852A (en) | Method for solving product assembly sequence planning problem on the basis of immune particle swarm algorithm | |
| Touairi et al. | Chaotic dynamics applied to piezoelectric harvester energy prediction with time delay | |
| CN113421621A (en) | Generation type countermeasure network driven silicon germanium super crystal lattice luminescent new material development technology | |
| Jiang et al. | Semiconductor manufacturing final test yield optimization and wafer acceptance test parameter inverse design using multi-objective optimization algorithms | |
| Jabi et al. | Graph machine learning using 3D topological models | |
| CN111444079A (en) | Software defect repairing method and device | |
| Amin et al. | A case-study led investigation of explainable AI (XAI) to support deployment of prognostics in industry | |
| Din et al. | Fuzzy adaptive teaching learning-based optimization strategy for GUI functional test cases generation | |
| Guidotti | Verification and repair of neural networks | |
| US10943182B2 (en) | Cognitive screening of EOR additives | |
| Suri et al. | Machine-learning based quantum cascade laser design | |
| Tian et al. | Web service reliability test method based on log analysis | |
| Gupta et al. | Randomized bias genetic algorithm to solve traveling salesman problem | |
| Akbar et al. | Machine learning approach to predicting tunnel field-effect transistors | |
| Ilunga et al. | Derivative-free Methods for Structural Optimization | |
| Dong et al. | HotBEV: Hardware-oriented transformer-based multi-view 3D detector for BEV perception | |
| Bozkurt et al. | Optimised realistic test input generation | |
| Butola et al. | Estimating the process variation effects of stacked gate all around si nanosheet CFETs using artificial neural network modeling framework |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |