Development of real-time density profile inversion with deep neural network model for multi-bands X-mode polarization reflectometer on EAST

Abstract Microwave reflectometry will serve as the primary diagnostic tool for measuring plasma density in ITER. The real-time inversion of high spatiotemporal resolution density profiles is a critical research focus for plasma fueling and control in both current and future fusion devices. A fast density profile inversion algorithm based on a deep neural network model has been proposed for the microwave reflectometry profile inversion database under various discharge parameters on EAST. Firstly, the model directly takes the high-sampling-rate raw time-domain <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mrow> <mml:mi>I</mml:mi> <mml:mrow> <mml:mo>/</mml:mo> </mml:mrow> <mml:mi>Q</mml:mi> </mml:mrow> </mml:math> data from the multi-bands (Q-, V- and W-bands) microwave reflectometer as input and demonstrates rapid and effective feature extraction capabilities. Secondly, the model outputs a two-dimensional vector containing both the positions of the microwave cut-off layers and the density information. Notably, the model exhibits excellent adaptability across different discharge parameters on EAST, including variations in magnetic field strength ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mrow> <mml:msub> <mml:mi>B</mml:mi> <mml:mrow> <mml:mi mathvariant="normal">t</mml:mi> <mml:mn>0</mml:mn> </mml:mrow> </mml:msub> </mml:mrow> </mml:math> = 1.57–2.30 T), different plasma confinement such as L-mode and H-mode, and during phases such as current ramp-up and steady-state discharge. The next step involves integrating this data-driven real-time density profile algorithm directly into the deployed profile reflectometer data acquisition system on EAST to enable parallel computation of data acquisition and processing. This integration aims to provide real-time density profile distribution, particularly during density feedback control experiments, such as gas or pellet injection experiments.

Authors

Institutions

Publication Details

Journal
Plasma Physics and Controlled Fusion
Published
2025-04-03
DOI
https://doi.org/10.1088/1361-6587/adc830
Citations
5
Primary Topic
Soil Moisture and Remote Sensing
Type
article
Field-Weighted Citation Impact
4.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Development of real-time density profile inversion with deep neural network model for multi-bands X-mode polarization reflectometer on EAST

Xiang Gao, Kaixuan Ye, Jia Huang, Gongshun Li et al.
5 citations
Plasma Physics and Controlled Fusion
Soil Moisture and Remote Sensing
4.00
article

Development of real-time density profile inversion with deep neural network model for multi-bands X-mode polarization reflectometer on EAST

Xiang Gao, Kaixuan Ye, Jia Huang, Gongshun Li, Chenguang Wan, Yan Feng, Zijie Liu, Kangning Geng, Fei Wen, Guoqiang Li, Tao Zhang, Zhen Zhou
article en
5 citations

Abstract

Abstract Microwave reflectometry will serve as the primary diagnostic tool for measuring plasma density in ITER. The real-time inversion of high spatiotemporal resolution density profiles is a critical research focus for plasma fueling and control in both current and future fusion devices. A fast density profile inversion algorithm based on a deep neural network model has been proposed for the microwave reflectometry profile inversion database under various discharge parameters on EAST. Firstly, the model directly takes the high-sampling-rate raw time-domain <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mrow> <mml:mi>I</mml:mi> <mml:mrow> <mml:mo>/</mml:mo> </mml:mrow> <mml:mi>Q</mml:mi> </mml:mrow> </mml:math> data from the multi-bands (Q-, V- and W-bands) microwave reflectometer as input and demonstrates rapid and effective feature extraction capabilities. Secondly, the model outputs a two-dimensional vector containing both the positions of the microwave cut-off layers and the density information. Notably, the model exhibits excellent adaptability across different discharge parameters on EAST, including variations in magnetic field strength ( <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mrow> <mml:msub> <mml:mi>B</mml:mi> <mml:mrow> <mml:mi mathvariant="normal">t</mml:mi> <mml:mn>0</mml:mn> </mml:mrow> </mml:msub> </mml:mrow> </mml:math> = 1.57–2.30 T), different plasma confinement such as L-mode and H-mode, and during phases such as current ramp-up and steady-state discharge. The next step involves integrating this data-driven real-time density profile algorithm directly into the deployed profile reflectometer data acquisition system on EAST to enable parallel computation of data acquisition and processing. This integration aims to provide real-time density profile distribution, particularly during density feedback control experiments, such as gas or pellet injection experiments.

Plasma Physics and Controlled FusionVol. 67(5)
University of Science and Technology of China (CN), Forschungszentrum Jülich (DE), Nanyang Technological University (SG), Chinese Academy of Sciences (CN), Institute of Plasma Physics (CN)
National Natural Science Foundation of China, Chinese Academy of Sciences, China Postdoctoral Science Foundation, Hefei Institutes of Physical Science, Chinese Academy of Sciences, National Postdoctoral Program for Innovative Talents, National Key Research and Development Program of China, Institute of Plasma Physics, Chinese Academy of Sciences
Openalex Percentile: Top 7%
Soil Moisture and Remote Sensing
4.00
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.