Advanced search
1 file | 2.30 MB Add to list

Real-time and adaptive reservoir computing with application to profile prediction in fusion plasma

Author
Organization
Abstract
Nuclear fusion is a promising alternative to address the problem of sustainable energy production. The tokamak is an approach to fusion based on magnetic plasma confinement, constituting a complex physical system with many control challenges. We study the characteristics and optimization of reservoir computing (RC) for real-time and adaptive prediction of plasma profiles in the DIII-D tokamak. Our experiments demonstrate that RC achieves comparable results to state-of-the-art (deep) convolutional neural networks (CNNs) and long short-term memory (LSTM) models, with a significantly easier and faster training procedure. This efficient approach allows for fast and frequent adaptation of the model to new situations, such as changing plasma conditions or different fusion devices.
Keywords
DISRUPTION PREDICTOR, REGRESSION, DESIGN, MODEL, JET, Plasmas, Reservoirs, Neurons, Adaptation models, Training, Biological, neural networks, Temperature measurement, Adaptive learning, condition, monitoring, nuclear fusion, reservoir computing (RC), tokamak plasma

Downloads

  • (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 2.30 MB

Citation

Please use this url to cite or link to this publication:

MLA
Jalalvand, Azarakhsh, et al. “Real-Time and Adaptive Reservoir Computing with Application to Profile Prediction in Fusion Plasma.” IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, vol. 33, no. 6, 2022, pp. 2630–41, doi:10.1109/TNNLS.2021.3085504.
APA
Jalalvand, A., Abbate, J., Conlin, R., Verdoolaege, G., & Kolemen, E. (2022). Real-time and adaptive reservoir computing with application to profile prediction in fusion plasma. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 33(6), 2630–2641. https://doi.org/10.1109/TNNLS.2021.3085504
Chicago author-date
Jalalvand, Azarakhsh, Joseph Abbate, Rory Conlin, Geert Verdoolaege, and Egemen Kolemen. 2022. “Real-Time and Adaptive Reservoir Computing with Application to Profile Prediction in Fusion Plasma.” IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 33 (6): 2630–41. https://doi.org/10.1109/TNNLS.2021.3085504.
Chicago author-date (all authors)
Jalalvand, Azarakhsh, Joseph Abbate, Rory Conlin, Geert Verdoolaege, and Egemen Kolemen. 2022. “Real-Time and Adaptive Reservoir Computing with Application to Profile Prediction in Fusion Plasma.” IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 33 (6): 2630–2641. doi:10.1109/TNNLS.2021.3085504.
Vancouver
1.
Jalalvand A, Abbate J, Conlin R, Verdoolaege G, Kolemen E. Real-time and adaptive reservoir computing with application to profile prediction in fusion plasma. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS. 2022;33(6):2630–41.
IEEE
[1]
A. Jalalvand, J. Abbate, R. Conlin, G. Verdoolaege, and E. Kolemen, “Real-time and adaptive reservoir computing with application to profile prediction in fusion plasma,” IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, vol. 33, no. 6, pp. 2630–2641, 2022.
@article{8755208,
  abstract     = {{Nuclear fusion is a promising alternative to address the problem of sustainable energy production. The tokamak is an approach to fusion based on magnetic plasma confinement, constituting a complex physical system with many control challenges. We study the characteristics and optimization of reservoir computing (RC) for real-time and adaptive prediction of plasma profiles in the DIII-D tokamak. Our experiments demonstrate that RC achieves comparable results to state-of-the-art (deep) convolutional neural networks (CNNs) and long short-term memory (LSTM) models, with a significantly easier and faster training procedure. This efficient approach allows for fast and frequent adaptation of the model to new situations, such as changing plasma conditions or different fusion devices.}},
  author       = {{Jalalvand, Azarakhsh and Abbate, Joseph and Conlin, Rory and Verdoolaege, Geert and Kolemen, Egemen}},
  issn         = {{2162-237X}},
  journal      = {{IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS}},
  keywords     = {{DISRUPTION PREDICTOR,REGRESSION,DESIGN,MODEL,JET,Plasmas,Reservoirs,Neurons,Adaptation models,Training,Biological,neural networks,Temperature measurement,Adaptive learning,condition,monitoring,nuclear fusion,reservoir computing (RC),tokamak plasma}},
  language     = {{eng}},
  number       = {{6}},
  pages        = {{2630--2641}},
  title        = {{Real-time and adaptive reservoir computing with application to profile prediction in fusion plasma}},
  url          = {{http://doi.org/10.1109/TNNLS.2021.3085504}},
  volume       = {{33}},
  year         = {{2022}},
}

Altmetric
View in Altmetric
Web of Science
Times cited: