TY - JOUR
T1 - Machine learning-based real-time kinetic profile reconstruction in DIII-D
AU - Shousha, Ricardo
AU - Seo, Jaemin
AU - Erickson, Keith
AU - Xing, Zichuan
AU - Kim, Sang Kyeun
AU - Abbate, Joseph
AU - Kolemen, Egemen
N1 - Publisher Copyright:
© 2023 The Author(s). Published by IOP Publishing Ltd on behalf of the IAEA. All rights reserved.
PY - 2024/2
Y1 - 2024/2
N2 - Kinetic equilibrium reconstruction plays a vital role in the physical analysis of plasma stability and control in fusion tokamaks. However, the traditional approach is subjective and prone to human biases. To address this, the consistent automatic kinetic equilibrium reconstruction (CAKE) method was introduced, providing objective results. Nonetheless, its offline nature limits its application in real-time plasma control systems (PCSs). To address this limitation, we present RTCAKENN, a machine learning model that approximates 7 CAKE-level output profiles, namely pressure, inverse q, toroidal current density, electron temperature and density, carbon ion impurity temperature and rotation profiles, using real-time available inputs. The deep neural network consists of an encoder layer, where the scalars and interdependent inputs such as plasma boundary coordinates and motional Stark effect data are encoded using multi-layer perceptrons (MLPs), while profile inputs are encoded by 1D convolutional layers. The encoded data is passed through a MLP for latent feature extraction, before being decoded in the decoding layers, which consist of upsampling and convolutional layers. RTCAKENN has been implemented in the DIII-D PCS and our model achieves accuracy comparable to CAKE and surpasses existing real-time alternatives. Through clever dropout training, RTCAKENN exhibits robustness and can operate even in the absence of Thomson scattering data or charge exchange recombination data. It executes in under 8 ms in the real-time environment, enabling future application in real-time control and analysis.
AB - Kinetic equilibrium reconstruction plays a vital role in the physical analysis of plasma stability and control in fusion tokamaks. However, the traditional approach is subjective and prone to human biases. To address this, the consistent automatic kinetic equilibrium reconstruction (CAKE) method was introduced, providing objective results. Nonetheless, its offline nature limits its application in real-time plasma control systems (PCSs). To address this limitation, we present RTCAKENN, a machine learning model that approximates 7 CAKE-level output profiles, namely pressure, inverse q, toroidal current density, electron temperature and density, carbon ion impurity temperature and rotation profiles, using real-time available inputs. The deep neural network consists of an encoder layer, where the scalars and interdependent inputs such as plasma boundary coordinates and motional Stark effect data are encoded using multi-layer perceptrons (MLPs), while profile inputs are encoded by 1D convolutional layers. The encoded data is passed through a MLP for latent feature extraction, before being decoded in the decoding layers, which consist of upsampling and convolutional layers. RTCAKENN has been implemented in the DIII-D PCS and our model achieves accuracy comparable to CAKE and surpasses existing real-time alternatives. Through clever dropout training, RTCAKENN exhibits robustness and can operate even in the absence of Thomson scattering data or charge exchange recombination data. It executes in under 8 ms in the real-time environment, enabling future application in real-time control and analysis.
KW - DIII-D
KW - equilibrium reconstruction
KW - kinetic
KW - machine learning
KW - neural net
KW - real-time
UR - http://www.scopus.com/inward/record.url?scp=85181094075&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85181094075&partnerID=8YFLogxK
U2 - 10.1088/1741-4326/ad142f
DO - 10.1088/1741-4326/ad142f
M3 - Article
AN - SCOPUS:85181094075
SN - 0029-5515
VL - 64
JO - Nuclear Fusion
JF - Nuclear Fusion
IS - 2
M1 - 026006
ER -