2021 Fiscal Year Research-status Report
Systematic Machine leARning Training for PORous mEdia characterizationS (SMART PORES)
| Project/Area Number |
21K13999
|
|---|
| Research Institution | Tokyo Institute of Technology |
|---|
Principal Investigator |
パトモノアジ アニンディティヨ 東京工業大学, 工学院, JSPS特別研究員 (20899265)
|
|---|
| Project Period (FY) |
2021-04-01 – 2023-03-31
|
| Keywords | Porous media / Machine Learning / Pore throat network / 3D printer / Microtomography |
|---|
| Outline of Annual Research Achievements |
Throughout this first year of research work, i have developed a method to incorporate 3D printer into this research work.
First, by using the 3D printer, I have successfully fabricated a custom 2D micromodel. This fabrication model is low cost and very useful for rapid prototyping. This result was published in an academic paper this year.
In addition, I have started 3D printing various granular particle that will be used for 3D porous media characterization. By using this artificially 3D printed granular particle, various porous media with various characteristics can be fabricated and further studies by using micro-tomography and image processing.
|
| Current Status of Research Progress |
Current Status of Research Progress
1: Research has progressed more than it was originally planned.
Reason
By incorporating a 3D printer, I can generate any particles that is required for the porous media characterization. Therefore, it will drastically increase the flexibility and open broader possibility to the scope of this research work because we can investigate various porous media in various applications.
In addition, comparison between these 3D printed particles and the naturally available particles can be performed to elucidate the difference between the simple and controlled particles to the more complicated shape of natural particles. As a result, the porous media characterization model for the artificially fabricated granular particles and naturally occurring granular particles can be made and thus compare to elucidate the differences.
|
| Strategy for Future Research Activity |
I will utilize the various generated particles to generate packed porous media. By using micro-tomography and image processing, I will derive various parameter, such as porosity, pore characteristics, and throat characteristics. By using generating the pore-throat networks, pore throat network simulations can be performed, and other parameters, such as permeability, can also be obtained. Afterward, I will develop the model for porous media characterization by using statistical and mathematical approach. Eventually, I will incorporate the machine learning by using the statistical data and also porous media image for the layer of deep learning.
|
| Causes of Carryover |
In this financial year, I have cancelled this funding because my recent affiliation is not eligible to receive this funding.
|
