Special Issue:
SPECIAL TOPIC — Plasma disruption
|
TOPICIAL REVIEW—Plasma disruption |
Prev
Next
|
|
|
Recent progress on deep learning-based disruption prediction algorithm in HL-2A tokamak |
Zongyu Yang(杨宗谕)1,†, Yuhang Liu(刘宇航)1, Xiaobo Zhu(朱晓博)1, Zhengwei Chen(陈正威)1, Fan Xia(夏凡)1, Wulyu Zhong(钟武律)1,‡, Zhe Gao(高喆)2, Yipo Zhang(张轶泼)1, and Yi Liu(刘仪)1 |
1 Southwestern Institute of Physics, Chengdu 610043, China; 2 Tsinghua University, Beijing 100084, China |
|
|
Abstract Disruption prediction and mitigation is a crucial topic, especially for future large-scale tokamaks, due to disruption's concomitant harmful effects on the devices. On this topic, disruption prediction algorithm takes the responsibility to give accurate trigger signal in advance of disruptions, therefore the disruption mitigation system can effectively alleviate the harmful effects. In the past 5 years, a deep learning-based algorithm is developed in HL-2A tokamak. It reaches a true positive rate of 92.2%, a false positive rate of 2.5% and a total accuracy of 96.1%. Further research is implemented on the basis of this algorithm to solve three key problems, i.e., the algorithm's interpretability, real-time capability and transferability. For the interpretability, HL-2A's algorithm gives saliency maps indicating the correlation between the algorithm's input and output by perturbation analysis. The distribution of correlations shows good coherence with the disruption causes. For the transferability, a preliminary disruption predictor is successfully developed in HL-2M, a newly built tokamak in China. Although only 44 shots are used as the training set of this algorithm, it gives reasonable outputs with the help of data from HL-2A and J-TEXT. For the real-time capacity, the algorithm is accelerated to deal with an input slice within 0.3 ms with the help of some adjustments on it and TFLite framework. It is also implemented into the plasma control system and gets an accuracy of 89.0% during online test. This paper gives a global perspective on these results and discusses the possible pathways to make HL-2A's algorithm a more comprehensive solution for future tokamaks.
|
Received: 18 January 2023
Revised: 03 April 2023
Accepted manuscript online: 07 April 2023
|
PACS:
|
52.35.Py
|
(Macroinstabilities (hydromagnetic, e.g., kink, fire-hose, mirror, ballooning, tearing, trapped-particle, flute, Rayleigh-Taylor, etc.))
|
|
52.55.Fa
|
(Tokamaks, spherical tokamaks)
|
|
07.05.Mh
|
(Neural networks, fuzzy logic, artificial intelligence)
|
|
52.55.-s
|
(Magnetic confinement and equilibrium)
|
|
Fund: Project supported by the National MCF R&D Program of China (Grant Nos. 2018YFE0302100 and 2019YFE03010003). The authors wish to thank all the members at South Western Institute of Physics for providing data, technique assistance and co-operating during the experiment. |
Corresponding Authors:
Zongyu Yang, Wulyu Zhong
E-mail: author:zy-yang@swip.ac.cn;zhongwl@swip.ac.cn
|
Cite this article:
Zongyu Yang(杨宗谕), Yuhang Liu(刘宇航), Xiaobo Zhu(朱晓博), Zhengwei Chen(陈正威), Fan Xia(夏凡), Wulyu Zhong(钟武律), Zhe Gao(高喆), Yipo Zhang(张轶泼), and Yi Liu(刘仪) Recent progress on deep learning-based disruption prediction algorithm in HL-2A tokamak 2023 Chin. Phys. B 32 075202
|
[1] Wesson J 2011 Tokamaks, fourth edition (Oxford University Press) pp. 15-27 [2] ITER Physics Expert Group on Disruptions 1999 Nucl. Fusion 39 2251 [3] Taylor P L, Kellman A G, Evans T E, et al 1999 Phys. Plasmas 6 1872 [4] Nardon E, Hu D, Hoelzl M, et al 2020 Nucl. Fusion 60 126040 [5] Vries P C, Johnson M F, Alper B, et al. 2011 Nucl. Fusion 51 053018 [6] Vries P C, Johnson M F, Alper B, et al. 2011 Nucl. Fusion 51 053018 [7] Rea C, Montes K J, Erickson K G, et al. 2019 Nucl. Fusion 59 096016 [8] Ratta G A, Vega J, Murari A, et al. 2010 Nucl. Fusion 50 025005 [9] Aymerich E, Sias G, Pisano F, et al. 2022 Nucl. Fusion 62 066005 [10] Zhong Y, Zheng, Chen Z Y, et al. 2021 Plasma Phys. Control. Fusion 63 075008 [11] Zheng W, Wu Q Q, Zhang M, et al. 2020 Plasma Phys. Control. Fusion 62 045012 [12] Guo B H, Chen D L, Shen B, et al. 2021 Plasma Phys. Control. Fusion 63 115007 [13] Harbeck J K, Svyatkovskiy A, Tang W, et al. 2019 Nature 568 526 [14] Zhu J X, Rea C, Montes K J, et al. 2021 Nucl. Fusion 61 026007 [15] Ma R and Wang A K 2010 Nucl. Fusion Plasma Phys. 30 37 [16] Yang Z Y, Xia F, Song X M, et al. 2020 Nucl. Fusion 60 016017 [17] Yang Z Y, Xia F, Song X M, et al. 2021 Nucl. Fusion 61 126042 [18] Yang Z Y, Xia F, Song X M, et al. 2022 Fusion Eng. Des. 182 113223 [19] Yang Z Y, Xia F, Song X M, et al. 2022 Proceeding on 48rm th EPS conference on plasma physics, June 27th-July 1st, 2022, Amsterdam, Holland, p4a.101 [20] Vries P C, Pautasso G, Humphreys D A, et al. 2016 Fusion Sci. Tech. 69 471 [21] Murari A, Rossi R, Peluso E, et al. 2020 Nucl. Fusion 60 056003 [22] Zhu J X, Rea C, Granetz R S, et al. 2021 Nucl. Fusion 61 114005 [23] Gelfusa M, Murari A, Vega J, et al. 2021 4th IAEA Technical Meeting on Fusion Data Processing, Validation and Analysis, November 29th-December 6th 2021, Chengdu, China, ID: 4 [24] Shen C S, Zheng W, Ding Y H, et al. 2023 Nucl. Fusion 63 046024 [25] Zhu J X, Rea C, Granetz R S, et al. 2023 Nucl. Fusion 63 046009 [26] Ganin Y, Ustinova E, Ajakan H, et al. 2016 Journal of Machine Learning Research 17 1 |
No Suggested Reading articles found! |
|
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
Altmetric
|
blogs
Facebook pages
Wikipedia page
Google+ users
|
Online attention
Altmetric calculates a score based on the online attention an article receives. Each coloured thread in the circle represents a different type of online attention. The number in the centre is the Altmetric score. Social media and mainstream news media are the main sources that calculate the score. Reference managers such as Mendeley are also tracked but do not contribute to the score. Older articles often score higher because they have had more time to get noticed. To account for this, Altmetric has included the context data for other articles of a similar age.
View more on Altmetrics
|
|
|