Researh on robust control of fusion plasma based on deep learing
| Project/Area Number |
17H03508
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Nuclear fusion studies
|
|---|
| Research Institution | The University of Tokyo |
|---|
Principal Investigator |
Ogawa Yuichi 東京大学, 大学院新領域創成科学研究科, 客員共同研究員 (90144170)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
日渡 良爾 国立研究開発法人量子科学技術研究開発機構, 六ヶ所核融合研究所 核融合炉システム研究開発部, 主幹研究員(定常) (40371348)
三善 悠矢 国立研究開発法人量子科学技術研究開発機構, 六ヶ所核融合研究所 核融合炉システム研究開発部, 任期付職員(任常) (50758638)
|
|---|
| Project Period (FY) |
2017-04-01 – 2020-03-31
|
| Project Status |
Completed (Fiscal Year 2019)
|
| Budget Amount *help |
¥16,380,000 (Direct Cost: ¥12,600,000、Indirect Cost: ¥3,780,000)
Fiscal Year 2019: ¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2018: ¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2017: ¥7,800,000 (Direct Cost: ¥6,000,000、Indirect Cost: ¥1,800,000)
|
|---|
| Keywords | 核融合 / プラズマ / トカマク / ディスラプション / 機械学習 / プラズマ・核融合 / 制御工学 |
|---|
| Outline of Final Research Achievements |
Data-driven science has been introduced for the foresight of plasma disruption which is one of the most important problems on tokamak plasma control including the ITER, paying attention to extracting dominant physical parameters among a lot of experimental data. By applying the Exhaust Search method in the machine learning technique for disruption prediction, a few key parameters (typically 5 ~ 7 ones) have been extracted though 23 physical parameters temporally and/or spatially measured.
In addition, using those parameters, the probability function of disruption occurrence has been derived, and the application to core plasma control has been proposed.
|
| Academic Significance and Societal Importance of the Research Achievements |
環境に優しく人類の恒久的なエネルギー源の有力な候補である核融合エネルギー開発は、国際プロジェクトITER計画を中心として精力的に推進されている。ITERをはじめとした将来の核融合炉心プラズマ制御の大きな課題として、プラズマが突然消滅するディスラプション現象があり、その制御に向けて実験や理論・シミュレーションで盛んに研究されている。
本研究では、最近急速に発展してきているビッグデータを活用したデータ駆動科学の手法をディスラプション予知に導入し、ディスラプションを引き起こす主要パラメータの抽出に成功すると共に、それらを用いたディスラプション制御の可能性を示した。
|
Report
(4 results)
Research Products
(10 results)
-
-
-
-
[Presentation] Applied study of feature extraction using exhaustive search on high-beta disruption in JT-60U2019
Author(s)
T. Yokoyama, Y. Miyoshi, R. Hiwatari, A. Isayama, G. Matsunaga, N. Oyama, Y. Igarashi, M. Okada, N. Imagawa, Y. Ogawa, and H. Yamada
Organizer
3rd Asia-Pacific Conference on Plasma Physics (AAPPS-DPP2019)
Related Report
Int'l Joint Research
-
[Presentation] Data-driven study of high-beta disruption prediction in JT-60U using exhaustive search2019
Author(s)
T. Yokoyama, Y. Miyoshi, R. Hiwatari, A. Isayama, G. Matsunaga, N. Oyama, Y. Igarashi, M. Okada, N. Imagawa, Y. Ogawa, and H. Yamada
Organizer
3rd Asia-Pacific Conference on Plasma Physics (AAPPS-DPP2019)
Related Report
Int'l Joint Research
-
[Presentation] Feature extraction using exhaustive search in disruption prediction based on JT-60U experimental data2019
Author(s)
T. Yokoyama, T. Sueyoshi, Y. Miyoshi, R. Hiwatari, A. Isayama, G. Matsunaga, N. Oyama, Y. Igarashi, M. Okada, N. Imagawa, H. Yamada, and Y. Ogawa
Organizer
2nd International Conference on Data Driven Plasma Science
Related Report
Int'l Joint Research
-
-
-
-
