Artificial Intelligence Empowering Digital Tokamak Systems: Research Progress, Challenges and Prospects

Chen Iv; Yifei Lv; Jing Ma

doi:10.31181/jidmgc11202516

Authors

Chen Iv Department of Information Technology Center, Southwestern Institute of Physics, China National Nuclear Corporation, China Author https://orcid.org/0009-0001-3489-2938
Yifei Lv Faculty of Economics and Management, ALFA University College, Selangor, Malaysia Author https://orcid.org/0009-0002-3593-2820
Jing Ma Department of Science, Technology and Informatization Southwestern Institute of Physics, China National Nuclear Corporation, China Author https://orcid.org/0009-0003-7138-8065

DOI:

https://doi.org/10.31181/jidmgc11202516

Keywords:

Artificial intelligence, Digital tokamak, Plasma control, Technical roadmap

Abstract

This study focuses on the application of artificial intelligence (AI) technologies in digital tokamak systems. Through systematic literature review and analysis, it expounds the application status, system construction, technical roadmap, and integration pathways of AI technologies in this field. The research shows that AI technologies have demonstrated significant advantages in plasma control, disruption prediction, and state recognition, effectively enhancing the performance and efficiency of digital tokamak systems. Meanwhile, this study constructs an architectural framework for AI-empowered digital tokamaks, and sorts out the technical roadmap covering multidisciplinary fields such as plasma physics, materials science, and control engineering, and proposes a full-process integration pathway from data fusion and model construction to application deployment. Although certain achievements have been made, challenges remain in model interpretability, data quality and scale, real-time requirements, and multi-scenario adaptability. In the future, deepening the application of AI in digital tokamak systems and advancing controlled nuclear fusion research to new heights will require interdisciplinary collaboration, algorithmic innovation, and data governance.

Downloads

Download data is not yet available.

References

Thomas, H. Stix. (1992). The elements of style (2nd ed.). American Institute of Physics. (Chapter 4).

Matsuoka, K., Miyamoto, K., Ohasa, K., & Wakatani, M. (1977). Magnetohydrodynamic instabilities in a current-carrying stellarator. Nuclear Fusion, 17(6), 1123-1132. https://doi.org/10.1088/0029-5515/17/6/002

Freidberg, J. P. (2014). Ideal magnetohydrodynamics (revised ed.). Cambridge University Press. https://doi.org/10.1017/CBO9780511795046

Manduchi, G., Luchetta, A., Taliercio, C., & Rigoni, A. (2018). Big data requirements in current and next fusion research experiments. 2018 IEEE International Symposium on Circuits and Systems (ISCAS), 1-5. https://doi.org/10.1109/ISCAS.2018.8351712

Chenguang, W., Zhi, Y., Feng, W., Xiaojuan, L., & Jiangang, Li. (2021). Experiment data-driven modeling of tokamak discharge in EAST. Nuclear Fusion, 61(6), 066015. https://doi.org/10.1088/1741-4326/abf419

Brendan, D. Tracey, Andrea, Michi., et al. (2024). Towards practical reinforcement learning for tokamak magnetic control. Fusion Engineering and Design, 200, 114161. https://doi.org/10.1016/j.fusengdes.2024.114161

Jeongwon, Lee., Jayhyun, Kim., et al. (2025). Machine learning based disruption prediction using long short-term memory in KSTAR. Nuclear Fusion, Accepted Manuscript online 3 July 2025. https://doi.org/10.1088/1741-4326/adeb9b

Zheng, W., Wu, Q. Q., & Zhang, M., et al. (2020). Disruption predictor based on neural network and anomaly detection on J-TEXT. Plasma Physics and Controlled Fusion, 62(4), 045012. https://doi.org/10.1088/1361-6587/ab6b02

Matos, F., Menkovski, V., Felici, F., Pau, A., Jenko, F., TCV Team, & EUROfusion MST1 Team. (2020). Classification of tokamak plasma confinement states with convolutional recurrent neural networks. Nuclear Fusion, 60(3), 036022. https://doi.org/10.1088/1741-4326/ab6c7a

Matos, F., Menkovski, V., & Pau, A., et al. (2021). Plasma confinement mode classification using a sequence-to-sequence neural network with attention. Nuclear Fusion, 61(4), 046019. https://doi.org/10.1088/1741-4326/ade3ed

Liu, G. Y., Peng, D. F., et al. (2021). Physical design of fusion target with edge computing. Physics: Conference Series, 2450, 012073. https://doi.org/10.1088/1742-6596/2450/1/012073

Li, H., Wang, L., Fu, Y. L., Wang, Z. X., Wang, T. B., & Li, J. Q. (2024). Surrogate model of turbulent transport in fusion plasmas using machine learning. Nuclear Fusion, 65(1), 016015. https://doi.org/10.1088/1741-4326/ad8b5b

Seo, J., Kim, S., Jalalvand, A., Conlin, R., Rothstein, A., Abbate, J., & Kolemen, E. (2024). Avoiding fusion plasma tearing instability with deep reinforcement learning. Nature, 626(8000), 746-751. https://doi.org/10.1038/s41586-024-07024-9

Katiyar, N., Goel, G., & Goel, S. (2021). Emergence of machine learning in the development of high entropy alloy and their prospects in advanced engineering applications. Emergent Materials, 4(6), 1635-1648. https://doi.org/10.1007/s42247-021-00249-8

Lupo Pasini, M., Samolyuk, G., Eisenbach, M., Choi, J. Y., Yin, J., & Yang, Y. (2024). First-principles data for solid solution niobium-tantalum-vanadium alloys with body-centered-cubic structures. Scientific Data, 11(1), 907. https://doi.org/10.1038/s41597-024-03720-3

Degrave, J., Felici, F., Buchli, J., Neunert, M., Tracey, B., Carpanese, F., & Riedmiller, M. (2022). Magnetic control of tokamak plasmas through deep reinforcement learning. Nature, 602(7897), 414-419. https://doi.org/10.1038/s41586-021-04301-9

Dubbioso, S., De Tommasi, G., Mele, A., Tartaglione, G., Ariola, M., & Pironti, A. (2023). A deep reinforcement learning approach for vertical stabilization of tokamak plasmas. Fusion Engineering and Design, 194, 113725. https://doi.org/10.1016/j.fusengdes.2023.113725

Litaudon, X., Jenko, F., Borba, D., Borodin, D. V., Braams, B. J., Brezinsek, S., & Zonca, F. (2022). EUROfusion-theory and advanced simulation coordination (E-TASC): programme and the role of high performance computing. Plasma Physics and Controlled Fusion, 64(3), 034005. https://doi.org/10.1088/1361-6587/ac44e4

Smith, S., Belli, E., Meneghini, O., Budiardja, R., Schissel, D., Candy, J., & Eubanks, A. (2022, August). A vision for coupling operation of US fusion facilities with HPC systems and the implications for workflows and data management. In Smoky Mountains Computational Sciences and Engineering Conference (pp. 87-100). Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-23606-8_6

Liu, Q. (2025). Research on distributed storage technology for next-generation tokamak experimental data [Doctoral dissertation, Huazhong University of Science and Technology]. https://doi.org/10.7666/d.D01546348

Stillerman, J. A., Mazon, D., Barbarino, M., Jain, P., de Sa, W. P., Xiao, B., & Nakanishi, H. (2025). Summary of the 14th IAEA Technical Meeting on Control Systems, Data Acquisition, Data Management and Remote Participation. Nuclear Fusion, 65(8), 087001. https://doi.org/10.1088/1741-4326/add605

Priyanka, M., Sangeetha, J., & Jayakumar, C. (2024). A review of traditional and data-driven approaches for disruption prediction in different tokamaks. In E3S Web of Conferences (Vol. 477, p. 00039). EDP Sciences. https://doi.org/10.1088/1361-6587/ac228b

Guo, B. H., Chen, D. L., Shen, B., & Rea, C., et al. (2021). Disruption prediction on EAST tokamak using a deep learning algorithm. Plasma Physics and Controlled Fusion, 63(11), 115007. https://doi.org/10.1088/1361-6587/ac228b

Tang, W., Feibush, E., Dong, G., & Borthwick, et al. (2024). AI-machine learning-enabled tokamak digital twin. arXiv preprint arXiv:2409.03112. https://doi.org/10.48550/arXiv.2409.03112

Battye, M. I., & Perinpanayagam, S. (2025). Digital twins in fusion energy research: Current state and future directions. IEEE Access, 13, 75787-75821. https://doi.org/10.1109/ACCESS.2025.3561920

Next Step Fusion. (n.d.). Plasma control solutions: From plasma shape controller to modern ML-based Plasma Control System (PCS). https://nextfusion.org/

Deng, G. H., Sun, Y., Wang, H. H., Shi, T., Gu, S., Ye, C., & Liu, H. (2025). Interpretability analysis and real-time prediction of locked mode-induced disruptions in EAST. Plasma Physics and Controlled Fusion, 67(7), 075013. https://doi.org/10.1088/1361-6587/ade5c5

Deng, G. H., Xie, P. G., & Sun, Y. W., et al. (2025). Automatic identification of tokamak plasma confinement states (L-mode, ELM-free H-mode, and ELMy H-mode) with multi-task learning neural network. Nuclear Fusion, 65(7), 076035. https://doi.org/10.1088/1741-4326/ade3ed

Yang, Z. Y. (2024). Application of artificial intelligence in the control and operation of HL-3 Tokamak. Report of 4TH SYMPOSIUM ON ARTIFICIAL INTELLIGENCE FOR CONTROLLED NUCLEAR FUSION. https://cj.sina.com.cn/articles/view/2343698037/8bb1fe7502701b6a0

Zhao, H. Y. (2024). Introduction to artificial intelligence in ENN Fusion. Report of 4TH SYMPOSIUM ON ARTIFICIAL INTELLIGENCE FOR CONTROLLED NUCLEAR FUSION. https://cj.sina.com.cn/articles/view/2343698037/8bb1fe7502701b6a0

Adadi, A., & Berrada, M. (2018). Peeking inside the black-box: A survey on explainable artificial intelligence (XAI). IEEE Access, 6, 52138-52160. https://doi.org/10.1109/ACCESS.2018.2870052

Lundberg, S. M., & Lee, S. I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 30, 4765-4774. https://proceedings.neurips.cc/paper/2017/hash/8a20a8621978632d76c43dfd28b67767-Abstract.html

Artificial Intelligence Empowering Digital Tokamak Systems: Research Progress, Challenges and Prospects

Authors

DOI:

Keywords:

Abstract

Downloads

References

Downloads

Published

Issue

Section

License

How to Cite

Cover page

template

The Journal is indexed or abstracted in:

Other information: