A5071056316- Magnetic confinement fusion research
- Ionosphere and magnetosphere dynamics
- Superconducting Materials and Applications
- Fusion materials and technologies
- Anomaly Detection Techniques and Applications
- Particle accelerators and beam dynamics
- Network Security and Intrusion Detection
- Plasma Diagnostics and Applications
- Nuclear Engineering Thermal-Hydraulics
- Nuclear reactor physics and engineering
- 3D IC and TSV technologies
- Laser-Plasma Interactions and Diagnostics
- Smart Grid Security and Resilience
- Laser-induced spectroscopy and plasma
- Solar and Space Plasma Dynamics
- Electronic Packaging and Soldering Technologies
- Advanced Fiber Optic Sensors
- Physics of Superconductivity and Magnetism
- Nuclear Physics and Applications
- Magnetic Field Sensors Techniques
- Magneto-Optical Properties and Applications
- Advanced Neural Network Applications
- Date Palm Research Studies
- Internet Traffic Analysis and Secure E-voting
- Advanced Database Systems and Queries
State Key Laboratory of Transducer Technology
2024-2025
Shanghai Institute of Microsystem and Information Technology
2024-2025
Beijing Institute of Technology
2024-2025
Hefei Institutes of Physical Science
2020-2024
Chinese Academy of Sciences
2015-2024
Institute of Plasma Physics
2015-2024
Sun Yat-sen University
2024
Huazhong University of Science and Technology
2007-2021
Beijing University of Technology
2019-2021
Université de Bordeaux
1995
Evidence of a nonlinear transition from mitigation to suppression the edge localized mode (ELM) by using resonant magnetic perturbations (RMPs) in EAST tokamak is presented. This first demonstration ELM with RMPs slowly rotating plasmas dominant radio-frequency wave heating. Changes topology after are indicated gradual phase shift plasma response field linear magneto hydro dynamics modeling result vacuum one and sudden increase three-dimensional particle flux divertor. The threshold depends...
This paper reports on disruption prediction using a shallow machine learning method known as random forest, trained large databases containing only plasma parameters that are available in real-time Alcator C-Mod, DIII-D, and EAST. The database for each tokamak contains sampled ∼106 times throughout ∼104 discharges (disruptive non-disruptive) over the last four years of operation. It is found number (e.g. , ) exhibit changes aggregate approached one or more these tokamaks. However, machine,...
A set of in-vessel resonant magnetic perturbation (RMP) coil has been recently installed in EAST. It can generate a range spectrum, and there is relatively large window for edge localized mode (ELM) control according to the vacuum field modeling island overlapping area. Observation mitigation suppression ELM slow rotating plasmas during application an n = 1 RMP presented this paper. Strong effect observed neutral beam injection heating plasmas. The frequency increases by factor 5, crash...
In this study, a long short-term memory (LSTM) model is trained on large disruption warning database to predict the EAST tokomak. To compare performance of proposed with previously reported full convolutional neural network (CNN) (Guo et al 2020 Plasma Phys. Control. Fusion 63 025008), same data set and diagnostic signals are used. Based test set, area under receiver operating characteristic curve, i.e. AUC value LSTM obtained as 0.87, true positive rate (TPR) sim87.5%, while false (FPR)...
Abstract Tokamaks are the most promising way for nuclear fusion reactors. Disruption in tokamaks is a violent event that terminates confined plasma and causes unacceptable damage to device. Machine learning models have been widely used predict incoming disruptions. However, future reactors, with much higher stored energy, cannot provide enough unmitigated disruption data at high performance train predictor before damaging themselves. Here we apply deep parameter-based transfer method...
The first equilibrium reconstruction of EAST current-density profile based on internal Faraday rotation measurements provided by the POlarimeter-INTerferometer (POINT) diagnostic is demonstrated using EFIT code. incorporates 11 simultaneous line-integrated density and effect from POINT to self-consistently reconstruct toroidal current a algorithm. It shown that can be applied improve accuracy core plasma q EAST. Comparisons magnetic surfaces reconstructed external data against those are...
Abstract A real-time disruption predictor using random forest was developed for high-density disruptions and used in the plasma control system (PCS) of EAST tokamak first time. The via (DPRF) ran piggyback mode actively exploited dedicated experiments during 2019–2020 experimental campaign to test its predictive capabilities oncoming disruptions. During experiments, mitigation triggered by a preset alarm provided DPRF neon gas injected into successfully mitigate damage. DPRF’s average...
Abstract Plasma disruption presents a significant challenge in tokamak fusion, especially large-size devices like ITER, where it causes severe damage. While current data-driven machine learning methods perform well prediction, they require extensive discharge data for model training. However, future tokamaks will begin operations without any prior data, making difficult to train predictors and select appropriate hyperparameters during the early operation period. In this period prediction...
Abstract Plasma disruption presents a significant challenge in tokamak fusion, especially large-size devices like ITER, where it causes severe damage. While current data-driven machine learning methods perform well prediction, they require extensive discharge data for model training. However, future tokamaks will begin operations without any prior data, making difficult to train predictors and select appropriate hyperparameters during the early operation period. In this period prediction...
This publisher’s note contains a correction to Opt. Lett. 50 , 241 ( 2025 ) 10.1364/OL.536982 .
Abstract In this study, a full convolutional neural network is trained on large database of experimental EAST data to classify disruptive discharges and distinguish them from non-disruptive discharges. The contains 14 diagnostic parameters the ∼10 4 (disruptive non-disruptive). test set 417 999 discharges, which are used evaluate performance model. results reveal that true positive (TP) rate ∼ 0.827, while false (FP) ∼0.067. This indicates 72 67 misclassified in set. FPs investigated detail...
Abstract Intrinsic error field on EAST is measured using the ‘compass scan’ technique with different n = 1 magnetic perturbation coil configurations in ohmically heated discharges. The intrinsic a non-resonant dominated spectrum even connection of upper and lower resonant coils order <?CDATA ${{b}_{r2,1}}/{{B}_{\text{T}}}\simeq {{10}^{-5}}$ ?> <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mstyle displaystyle="false"> <mml:msub> <mml:mrow> <mml:mi>b</mml:mi>...
Next generation high performance (HP) tokamaks risk damage from unmitigated disruptions at current and power. Achieving reliable disruption prediction for a device's HP operation based on its low (LP) data is key to success. In this letter, through explorative analysis dedicated numerical experiments multiple existing tokamaks, we demonstrate how the operational regimes of can affect power trained predictor. First, our results suggest data-driven predictors abundant LP discharges work poorly...
Abstract Plasma disruption is a very dangerous event for future tokamaks and fusion reactors. Therefore, predicting crucial ensuring the safety performance of In this study, features two deep learning algorithms are integrated to establish multi-scale hybrid network predictor. Firstly, 43 diagnostic signals extracted by convolutional neural (CNN), time information learned long short-term memory network. The predictor trained tested on database containing <?CDATA ${\sim}10^4$?> <mml:math...
Abstract Since the last IAEA-FEC in 2021, significant progress on development of long pulse steady state scenario and its related key physics technologies have been achieved, including reproducible 403 s long-pulse steady-state H-mode plasma with pure radio frequency (RF) power heating. A thousand-second time scale (∼1056 s) fully non-inductive high injected energy up to 1.73 GJ has also achieved. The EAST operational regime β P significantly extended ( H 98y2 > 1.3, ∼ 4.0, N 2.4 n e / GW...
A toroidal Alfvén eigenmode (TAE) excited by barely trapped energetic electrons during the application of a static magnetic perturbations (MPs) is observed for first time in tokamak ohmic heating plasmas. This TAE appears when current n = 2 MPs exceeds threshold value, at which forced reconnection happens. Here, mode number. located near plasma edge, agrees with calculation gap. It propagates ion diamagnetic direction and has dominant number 2. The frequency consistent precessional energy...
A preliminary analysis of plasma current quenching is presented in this paper based on the disruption database. It demonstrates that 26.8% discharges have been disrupted last 2012 campaign, addition, disruptive rate grows with increase current. The best-fit linear and instantaneous quench extracted from recent EAST disruptions, showing an 80%–30% interval maximum well fit for device. lowest area-normalized time 3.33 ms/m2 estimated electron temperature being 7.3 eV,~9.5 eV. In case eddy goes...
High pressure noble gas injection is a promising technique to mitigate the effect of disruptions in tokamaks. In this paper, results mitigation experiments with low-Z massive (helium) on EAST tokamak are reported. A fast valve has been developed and successfully implemented EAST, response time ⩽150 μs, capable injecting up particles, corresponding 300 times plasma inventory.
Rising harvest costs and looming labor shortages are threatening profitability in the sweet cherry (Prunus avium L.) industry. The current study obtained baseline data on performance efficiency for potential mechanical or mechanically-assisted systems improving of fresh market grade cherries. We compared a prototype system that detaches fruit using impact force, with vibration, through series excited vibration tests. A group single axial accelerometers were installed at different locations...
The method of plasma current profile reconstruction using the polarimeter/interferometer (POINT) data from a simulated equilibrium is explored and validated. It shown that safety factor (q) can be generally reconstructed external magnetic POINT data. q found to reasonably agree with initial equilibriums. Comparisons density profiles 3%, 5% 10% random errors are investigated. result shows could used accurate determination profile.