FusionPapers
図版検索AI要約wiki日本の研究装置ジャーナルChatGPTAbout
© 2026 FUSIONPAPERS
About法務情報
トップに戻る

Physics-informed deep learning model for line-integral diagnostics across fusion devices

Cong Wang, Weizhe Yang, Haiping Wang, Renjie Yang, Jing Li, Zhijun Wang, Yixiong Wei, Xianli Huang, Chenshu Hu, Zhaoyang Liu2025年7月Nuclear FusionIF 3出版社

Rapid reconstruction of 2D plasma profiles from line-integral measurements is important in nuclear fusion. This paper introduces a physics-informed model architecture called Onion, that can enhance the performance of models and be adapted to various backbone networks. The model under Onion incorporates physical information (PI) by a multiplication process and applies the physics-informed loss function (PILF) according to the principle of line integration. Prediction results demonstrate that the additional input of PI improves the deep learning model’s ability, leading to a reduction in the average relative error between the reconstruction profiles and the target profiles by approximately on synthetic datasets and about on experimental datasets. Furthermore, the implementation of the Softplus activation function in the final two fully connected layers improves model performance. This enhancement results in a reduction in the by approximately on synthetic datasets and about on experimental datasets. The incorporation of the PILF has been shown to correct the model’s predictions, bringing the back-projections closer to the actual inputs and reducing the errors associated with inversion algorithms. Besides, we have developed a synthetic data model to generate customized line-integral diagnostic datasets and have also collected soft x-ray diagnostic datasets from EAST and HL-2A. This study achieves reductions in reconstruction errors, and accelerates the development of surrogate models in fusion research.

装置

east中精度(概要文一致)

wiki

Plasma diagnosticsDeep learning

AIによる論文要約

プラズマ診断のための物理情報深層学習モデル
JAこの論文は、核融合プラズマ診断に関心のある研究者や学生に役立つでしょう。特に、物理情報を組み込んだ深層学習モデルの開発や、線積分測定データの利用に興味のある人に向けています。#核融合 #プラズマ診断 #物理情報深層学習 #線積分測定 #サロゲートモデル
LLM向け: {'Title': 'プラズマ診断のための物理情報深層学習モデル', 'Author(s)': '不明', 'Research Objective': '線積分…

この論文は、線積分測定から2Dプラズマプロファイルを迅速に再構築するための物理情報深層学習モデルを紹介しています。物理情報を組み込むことで、合成および実験データセットの再構築誤差を大幅に低減できることを示しています。この研究は、核融合研究におけるサロゲートモデルの開発を加速させます。

Physics-informed deep learning model for line-integral diagnostics across fusion devices
ENThis paper should be read by fusion researchers and engineers interested in developing advanced diagnostic techniques for plasma profile reconstruction, as well as those working on the application of deep learning in fusion research.#FusionDiagnostics #PlasmaProfileReconstruction #PhysicsInformedDeepLearning
LLM向け: {'Title': 'Physics-informed deep learning model for line-integral diagnostics ac…

This paper presents a physics-informed deep learning model called Onion that can accurately reconstruct 2D plasma profiles from line-integral measurements. The model incorporates physical information and a physics-informed loss function to improve performance, leading to a significant reduction in reconstruction errors on both synthetic and experimental datasets.

関連論文

Data reconstruction using variational autoencoders and error analysis compared to B-spline interpolation

2026Plasma Physics and Controlled Fusion

Multi-diagnostics reconstruction of magnetic equilibrium and kinetic profiles using physics-informed neural networks with applications to JET

2026Nuclear Fusion

Efficient training sets for surrogate models of tokamak turbulence with Active Deep Ensembles

2024Nuclear Fusion

Simultaneous kinetic profile and magnetic equilibrium inference with Bayesian integrated data analysis in preparation for ITER

2025Plasma Physics and Controlled Fusion

High-fidelity and explainable machine learning method for neoclassical toroidal viscosity torque prediction

2026Nuclear Fusion

Summary report of the 3rd IAEA technical meeting on fusion data processing validation and analysis (FDPVA)

2020Nuclear Fusion

Deep learning based plasma response models to 3D external magnetic field perturbations in EAST

2026Nuclear Fusion

Physics-informed neural networks for the modelling of interferometer-polarimetry in tokamak multi-diagnostic equilibrium reconstructions

2025Plasma Physics and Controlled Fusion

Bayesian modelling of fusion diagnostics

2003Plasma Physics and Controlled Fusion

Neural network approximation of Bayesian models for the inference of ion and electron temperature profiles at W7-X

2019Plasma Physics and Controlled Fusion