2020 Fiscal Year Final Research Report
Estimation of rheological characteristics of dual phase fluid at high temperature utilizing transfer learning
| Project/Area Number |
18H01762
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 26060:Metals production and resources production-related
|
|---|
| Research Institution | Kyushu University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
西郷 浩人 九州大学, システム情報科学研究院, 准教授 (90586124)
|
|---|
| Project Period (FY) |
2018-04-01 – 2021-03-31
|
| Keywords | 見かけ粘度 / サスペンション / フォーム / 非ニュートン流体 / 粘弾性流体 / 機械学習 / ガウス過程 |
|---|
| Outline of Final Research Achievements |
In the present study, the viscosity of suspensions of dispersed particles consisting of polyethylene beads in a matrix of silicone oil or aqueous glycerol at room temperature was measured. Then, empirical models for estimating the viscosity were proposed. Furthermore, the viscosity of suspensions of CaO and MgO particles dispersed in a matrix of CaO-Al2O3-SiO2-MgO slag at 1773 K was measured, and the feasibility of the proposed viscosity equations was investigated. Under comparable measurement conditions, the viscosities of the glycerol suspensions were higher than those of the silicone oil suspensions. The trend of increasing viscosity of the molten slag suspensions was similar to that of the room-temperature suspensions, exhibiting Bingham non-Newtonian behavior. The viscosity model composed with the results from the glycerol aqueous suspensions underestimated the slag viscosity, which can be attributed to the repulsive forces in the high-polarity liquid matrix.
|
| Free Research Field |
高温融体物理化学
|
| Academic Significance and Societal Importance of the Research Achievements |
本課題では，研究代表者がこれまで開発してきた高温二相共存融体（サスペンションおよびフォーム）の粘性に関する一連の手法を余すことなく駆使し，高精度な実験データを積み上げた．ここでは，比較的に実験が容易でデータ数の大きな室温系の見かけ粘度挙動から，機械学習に関する優れた実績が豊富な研究分担者とともに，挙動が似て非なる高温系の流動挙動を推定するモデルを転移学習および特徴選択によって構築し，同分野における機械学習の適用を世界に先駆けて行った．また，得られたモデルはエクセル形式のスプレッドシートとして，国内の金属素材産業に携わる研究者に試用をしてもらっている．
|
