Development of real-time ELM control system based on deep learning
 in HL-2A tokamak

doi:10.16568/j.0254-6086.202304002

Nuclear Fusion and Plasma Physics ›› 2023, Vol. 43 ›› Issue (4): 380-385.DOI: 10.16568/j.0254-6086.202304002

• Nuclear Fusion Engineering • Previous Articles Next Articles

Development of real-time ELM control system based on deep learning in HL-2A tokamak

LI Xuan-yi, XIA Fan, YANG Zong-yu, GONG Xin-wen, CHEN Cheng-yuan, XIAO Guo-liang, ZHU Xiao-bo

(Southwestern Institute of Physics, Chengdu 610041)

Received:2021-12-03 Revised:2023-05-22 Online:2023-12-15 Published:2023-12-06

基于深度学习的 HL-2A 装置 ELM 实时控制系统的研制

李宜轩，夏凡，杨宗谕，龚新文，陈程远，肖国梁，朱晓博

(核工业西南物理研究院，成都 610041)

作者简介:李宜轩(1996-)，男，辽宁抚顺人，硕士研究生，从事等离子控制研究。
基金资助:
国家研发计划(2018YFE0302100)；国家自然科学基金(11875022)

Abstract

Abstract: A real-time ELM (edge localized mode) control and mitigation system was developed in HL-2A tokamak using the deep learning-based ELM identification algorithm and ELM mitigation equipment such as supersonic molecular beam injection (SMBI). The system collects relevant input data in real time, calculates and analyzes them with the neural network, and sends out trigger signals according to the output of neural network. When continuous occurrence of ELM is detected, the system will trigger SMBI to inject molecular beams to mitigate ELM. The real-time ELM control system was tested in the HL-2A tokamak experiment. It can accurately detect the ELMy H mode in the plasma discharge by analyzing the diagnostic signal within a control period of 1ms and sends control signals when ELM occurs to trigger the ELM mitigation system, and the identification effect is obvious. The system can be applied to the real-time ELM mitigation and control in HL-2A plasma.

Key words: ELM, PCS, H-mode, Real-time ELM system

摘要： 利用 HL-2A 装置已开发的基于深度学习的边缘局域模 (ELM)识别算法和超声分子束注入(SMBI)等 ELM 缓解设备组成了一个 ELM 实时识别和控制系统。该系统实时采集相关的输入数据，通过神经网络计算分析，输出识别信号，当检测到存在连续 ELM 时，触发 SMBI 以缓解 ELM。在 HL-2A 装置放电实验期间对 ELM 实时控制系统进行了测试，识别效果明显，在 1ms 控制周期中，达到了 ELM 的实时缓解与控制。

关键词: 边缘局域模, 等离子体控制系统, H 模, ELM 实时控制系统

CLC Number:

TL62+ 9.1

LI Xuan-yi, XIA Fan, YANG Zong-yu, GONG Xin-wen, CHEN Cheng-yuan, XIAO Guo-liang, ZHU Xiao-bo. Development of real-time ELM control system based on deep learning in HL-2A tokamak[J]. Nuclear Fusion and Plasma Physics, 2023, 43(4): 380-385.

李宜轩, 夏凡, 杨宗谕, 龚新文, 陈程远, 肖国梁, 朱晓博 . 基于深度学习的 HL-2A 装置 ELM 实时控制系统的研制[J]. 核工业西南物理研究院, 2023, 43(4): 380-385.

References

[1] Wagner F, Becker G, Behringer K, et al. Regime of improved confinement and high beta in neutral-beamheated divertor discharges of the ASDEX tokamak [J]. Phys. Rev. Lett., 1982, 49(19): 1408-1412.

[2] Jahns G L, Groebner R J, John H S. Comparison of transport in H- and L-phase discharges in the DIII-D tokamak [J]. Nucl. Fusion, 1989, 29(8): 1271.

[3] ASDEX Team. The H-mode of ASDEX [J]. Nucl. Fusion, 1989, 29(11): 1959−2040.

[4] Zohm H, Wagner F, Endler M, et al. Studies of edge localized modes on ASDEX [J]. Nucl. Fusion, 1992, 32(3): 489-494.

[5] Wang L, Xu G S, Guo H Y, et al. Particle and power deposition on divertor targets in EAST H-mode plasmas [J]. Nucl. Fusion, 2012, 52(6): 063024.

[6] Eich T, Andrew P, Herrmann A, et al. ELM resolved energy distribution studies in the JET MKII gas-box divertor using infra-red thermography [J]. Plasma Phys. Contr. Fusion, 2007, 49(5): 573−604.

[7] Petrie T W, Evans T E, Brooks N H, et al. Results from radiating divertor experiments with RMP ELM suppression and mitigation [J]. Nucl. Fusion, 2011, 51(7): 073003.

[8] Thornton A J, Kirk A, Cahyna P, et al. The effect of resonant magnetic perturbations on the divertor heat and particle fluxes in MAST [J]. Nucl. Fusion, 2014, 54(6): 064011.

[9] Lang P T, Neuhauser J, Horton L D, et al. ELM frequency control by continuous small pellet Injection in ASDEX Upgrade [J]. Nucl. Fusion, 2003, 43: 1110-1120.

[10] Baylor L R, Commaux N, Jernigan T C, et al. Reduction of edge-localized mode intensity using high-repetition -rate pellet injection in tokamak H-mode plasmas [J]. Phys. Rev. Lett., 2013, 110(24): 245001.

[11] Rapp J, Garbet P M, Matthews G F, et al. Reduction of divertor heat load in JET ELMy H-modes using impurity seeding techniques [J]. Nucl. Fusion, 2004, 44: 312−319.

[12] Matos F A, Ferreira D R, Carvalho P J. Deep learning for plasma tomography using the bolometer system at JET [J]. Fusion Eng. Design, 2017, 114: 18−25.

[13] Rietman E A, Beachy M. A study on failure prediction in a plasma reactor [J]. IEEE Transactions on Semiconductor Manufacturing, 1998, 11(4): 670−680.

[14] Cannas B, Cau F, Fanni A, et al. Automatic disruption classification at JET: comparison of different pattern recognition techniques [J]. Nucl. Fusion, 2008, 46(7): 699.

[15] 刘春华, 侯智培, 王瑜琴, 等. 人工神经网络在 HL-2A 装置汤姆逊散射数据处理中的应用 [J]. 强激光与粒子束, 2019, 31(2): 41−47.

[16] 黄尧，夏凡，杨宗谕, 等. 基于深度学习的 ELM 实时识别研究 [J]. 核聚变与等离子体物理, 2020, 40(4): 301−308.

[17] 高金明, 程钧, 严龙文, 等. HL-2A 装置超声分子束注入缓解偏滤器靶板上边缘局域模热通量研究 [J]. 核聚变与等离子体物理, 2015, 35(1): 1−7.

[18] Yao L H, Tang N Y, Cui Z Y, et al. Plasma behavior with molecular beam injection in the HL-1M tokamak [J]. Nucl. Fusion, 1998, 38(4): 631-638.

[19] 冯北滨, 姚良骅, 李伟, 等. HL-2A 装置偏滤器位形超声分子束注入深度观测 [J]. 核聚变与等离子体物理, 2008, 28(2): 97-100.

[20] Yu D L, Chen C Y, Yao L H, et al. Penetration characteristics of supersonic molecular beam injection on HL-2A tokamak [J]. Nucl. Fusion, 2010, 50(3): 035009.

Development of real-time ELM control system based on deep learning in HL-2A tokamak

基于深度学习的 HL-2A 装置 ELM 实时控制系统的研制

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 3

Recommended Articles

Metrics

[1]	HUANG Yao, XIA Fan, YANG Zong-yu, ZHONG Wu-lü, LIU Chun-hua . ELM real-time recognition research based on deep learning [J]. NUCLEAR FUSION AND PLASMA PHYSICS, 2020, 40(4): 300-308.
[2]	PEI Xiao-fang, YUAN Qi-ping, HUANG Yao, LUO Zheng-ping, XIAO Bing-jia. Integration of real-time GPU parallel equilibrium reconstruction in EAST plasma control system [J]. NUCLEAR FUSION AND PLASMA PHYSICS, 2017, 37(1): 105-112.
[3]	GAO Xiao-yan, TIAN Pei-hong, CUI Cheng-he, CAI Xiao. Plasma density feedback control system in HL-2A tokamak [J]. NUCLEAR FUSION AND PLASMA PHYSICS, 2012, 32(2): 158-164.