Adaptive high learning rate probabilistic disruption predictors from scratch for the next generation of tokamaks

The development of accurate real-time disruption predictors is a pre-requisite to any mitigation action. Present theoretical models of disruptions do not reliably cope with the disruption issues. This article deals with data-driven predictors and a review of existing machine learning techniques, from both physics and engineering points of view, is provided. All these methods need large training datasets to develop successful predictors. However, ITER or DEMO cannot wait for hundreds of disruptions to have a reliable predictor. So far, the attempts to extrapolate predictors between different tokamaks have not shown satisfactory results. In addition, it is not clear how valid this approach can be between present devices and ITER/DEMO, due to the differences in their respective scales and possibly underlying physics. Therefore, this article analyses the requirements to create adaptive predictors from scratch to learn from the data of an individual machine from the beginning of operation. A particular algorithm based on probabilistic classifiers has been developed and it has been applied to the database of the three first ITER-like wall campaigns of JET (1036 non-disruptive and 201 disruptive discharges). The predictions start from the first disruption and only 12 re-trainings have been necessary as a consequence of missing 12 disruptions only. Almost 10 000 different predictors have been developed (they differ in their features) and after the chronological analysis of the 1237 discharges, the predictors recognize 94% of all disruptions with an average warning time (AWT) of 654 ms. This percentage corresponds to the sum of tardy detections (11%), valid alarms (76%) and premature alarms (7%). The false alarm rate is 4%. If only valid alarms are considered, the AWT is 244 ms and the standard deviation is 205 ms. The average probability interval about the reliability and accuracy of all the individual predictions is 0.811 ± 0.189.

[1]  David G. Stork,et al.  Pattern Classification , 1973 .

[2]  K. F. Mast,et al.  Disruptions in JET , 1989 .

[3]  Teuvo Kohonen,et al.  The self-organizing map , 1990 .

[4]  M. F. F. Nave,et al.  Mode locking in tokamaks , 1990 .

[5]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[6]  Paul M. Frank,et al.  Fault Diagnosis in Dynamic Systems , 1993, Robotics, Mechatronics and Manufacturing Systems.

[7]  Christopher M. Bishop,et al.  GTM: The Generative Topographic Mapping , 1998, Neural Computation.

[8]  Vladimir Cherkassky,et al.  Learning from data , 1998 .

[9]  Alexander Gammerman,et al.  Machine-Learning Applications of Algorithmic Randomness , 1999, ICML.

[10]  Alexander Gammerman,et al.  Transduction with Confidence and Credibility , 1999, IJCAI.

[11]  A. Sengupta,et al.  Forecasting disruptions in the ADITYA tokamak using neural networks , 2000 .

[12]  Shigeo Abe DrEng.  Pattern Classification , 2001, Springer London.

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

[14]  Harry Wechsler,et al.  Transductive confidence machine for active learning , 2003, Proceedings of the International Joint Conference on Neural Networks, 2003..

[15]  Jet Efda Contributors,et al.  A cross-tokamak neural network disruption predictor for the JET and ASDEX Upgrade tokamaks , 2005 .

[16]  W. Gasarch,et al.  The Book Review Column 1 Coverage Untyped Systems Simple Types Recursive Types Higher-order Systems General Impression 3 Organization, and Contents of the Book , 2022 .

[17]  Piergiorgio Sonato,et al.  A prediction tool for real-time application in the disruption protection system at JET , 2007 .

[18]  J. Manickam,et al.  Chapter 3: MHD stability, operational limits and disruptions , 2007 .

[19]  Zhiyuan Luo,et al.  Reliable Probabilistic Classification and Its Application to Internet Traffic , 2008, ICIC.

[20]  Harry Wechsler,et al.  Query by Transduction , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[21]  Luigi Fortuna,et al.  Prototype of an adaptive disruption predictor for JET based on fuzzy logic and regression trees , 2008 .

[22]  P. Arena,et al.  A Disruption Predictor Based on Fuzzy Logic Applied to JET Database , 2008, IEEE Transactions on Plasma Science.

[23]  Sethuraman Panchanathan,et al.  Generalized Query by Transduction for online active learning , 2009, 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops.

[24]  G. Granucci,et al.  Disruption control on FTU and ASDEX upgrade with ECRH , 2009 .

[25]  J. Contributors,et al.  Statistical analysis of disruptions in JET , 2009 .

[26]  F. Saint-Laurent,et al.  Experimental study of disruption mitigation using massive injection of noble gases on Tore Supra , 2010 .

[27]  D. A. Humphreys,et al.  Demonstration of rapid shutdown using large shattered deuterium pellet injection in DIII-D , 2010 .

[28]  B. Cannas,et al.  An adaptive real-time disruption predictor for ASDEX Upgrade , 2010 .

[29]  Mohammad Bakhtiari,et al.  Using mixed gases for massive gas injection disruption mitigation on Alcator C-Mod , 2010 .

[30]  Andrea Murari,et al.  An advanced disruption predictor for JET tested in a simulated real-time environment , 2010 .

[31]  Alessandra Fanni,et al.  Disruption prediction with adaptive neural networks for ASDEX Upgrade , 2011 .

[32]  P. D. Morgan,et al.  Disruption mitigation by massive gas injection in JET , 2011 .

[33]  J. Contributors,et al.  Survey of disruption causes at JET , 2011 .

[34]  D. A. Humphreys,et al.  Novel rapid shutdown strategies for runaway electron suppression in DIII-D , 2011 .

[35]  G. Pautasso,et al.  Prediction of disruptions on ASDEX Upgrade using discriminant analysis , 2011 .

[36]  J. Vega,et al.  Results of the JET real-time disruption predictor in the ITER-like wall campaigns , 2012 .

[37]  Vladimir Vovk,et al.  Multiprobabilistic Venn Predictors with Logistic Regression , 2012, AIAI.

[38]  DIII-D Team,et al.  The limits and challenges of error field correction for ITERa) , 2012 .

[39]  Allen H. Boozer,et al.  Theory of tokamak disruptionsa) , 2012 .

[40]  R. H. Bulmer,et al.  Sustained Spheromak Physics Experiment (SSPX): design and physics results , 2012 .

[41]  Harris Papadopoulos,et al.  Reliable Probability Estimates Based on Support Vector Machines for Large Multiclass Datasets , 2012, AIAI.

[42]  R. Neu,et al.  The impact of the ITER-like wall at JET on disruptions , 2012 .

[43]  Jet Efda Contributors,et al.  Development of an efficient real-time disruption predictor from scratch on JET and implications for ITER , 2013 .

[44]  Alessandra Fanni,et al.  Multivariate statistical models for disruption prediction at ASDEX Upgrade , 2013 .

[45]  S. A. Sabbagh,et al.  Detection of disruptions in the high-β spherical torus NSTX , 2013 .

[46]  Harris Papadopoulos,et al.  Reliable probabilistic classification with neural networks , 2013, Neurocomputing.