Detection and prediction of a beam-driven mode in field-reversed configuration plasma with recurrent neural networks

Energetic beams excite semi-repetitive modes ('staircase mode') in the field-reversed configuration (FRC) plasma. We explore several neural network architectures to detect, and in some cases predict, this type of mode onset. We weigh the performance of these architectures and find that recurrent neural networks (RNNs), specifically long short-term memory (LSTM) networks, outperform all other models we examine. LSTMs can predict the onset of staircase with a lead window of 0.2 ms, which has implications for plasma longevity and is a promising direction for similar analysis in FRC devices in the future.

[1]  Lovekesh Vig,et al.  Long Short Term Memory Networks for Anomaly Detection in Time Series , 2015, ESANN.

[2]  D. Q. Bui,et al.  Formation of a long-lived hot field reversed configuration by dynamically merging two colliding high-β compact toroidsa) , 2011 .

[4]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[5]  I. Jolliffe.  Principal Component Analysis , 2002 .

[6]  S. Putvinski,et al.  Diagnostic suite of the C-2U advanced beam-driven field-reversed configuration plasma experiment. , 2016, The Review of scientific instruments.

[7]  T. Tajima,et al.  Forecast of TEXT plasma disruptions using soft X rays as input signal in a neural network , 1999 .

[8]  Sean Dettrick,et al.  A high performance field-reversed configurationa) , 2015 .

[9]  Karol J. Piczak.  Environmental sound classification with convolutional neural networks , 2015, 2015 IEEE 25th International Workshop on Machine Learning for Signal Processing (MLSP).

[10]  Lovekesh Vig,et al.  Anomaly detection in ECG time signals via deep long short-term memory networks , 2015, 2015 IEEE International Conference on Data Science and Advanced Analytics (DSAA).

[11]  K. F. Mast,et al.  On-line prediction and mitigation of disruptions in ASDEX Upgrade , 2002 .

[12]  Y. Luo,et al.  Magnetic diagnostic suite of the C-2 field-reversed configuration experiment confinement vessel. , 2012, The Review of scientific instruments.

[13]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[15]  Wendell Horton,et al.  Neural network prediction of some classes of tokamak disruptions , 1996 .

[16]  R. M. Magee,et al.  Formation of hot, stable, long-lived field-reversed configuration plasmas on the C-2W device , 2019, Nuclear Fusion.

[17]  Shai Ben-David,et al.  Understanding Machine Learning: From Theory to Algorithms , 2014 .

[18]  Nasser M. Nasrabadi,et al.  Pattern Recognition and Machine Learning , 2006, Technometrics.

[19]  Sebastián Dormido-Canto,et al.  Automatic recognition of anomalous patterns in discharges by recurrent neural networks , 2020 .

[20]  David Haussler,et al.  Quantifying Inductive Bias: AI Learning Algorithms and Valiant's Learning Framework , 1988, Artif. Intell..

[21]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[22]  R. M. Magee,et al.  First experimental measurements of a new fast ion driven micro-burst instability in a field-reversed configuration plasma , 2018, Nuclear Fusion.

[23]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[24]  T. Schindler,et al.  Integrated diagnostic and data analysis system of the C-2W advanced beam-driven field-reversed configuration plasma experiment. , 2018, The Review of scientific instruments.

[25]  Alessandra Fanni,et al.  Disruption forecasting at JET using neural networks , 2002 .

[26]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[27]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[28]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[29]  A. Ivanov,et al.  Low energy, high power hydrogen neutral beam for plasma heating. , 2015, The Review of scientific instruments.

[30]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.