YCYihang Chen

Yihang Chen

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yihang Chen has a total of 162 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    First experimental observation of ‘Staircase’ high confinement mode in tokamak plasma

    by Y. Zhang, Z.B. Guo, X.G. Wang, C. Li, Y.R. Zhu, G.Q. Dong, Y.F. Wang, Z.Y. Li, M.J. Choi, Z.J. Li, et al.

    In magnetically confined nuclear fusion, the achievement of ignition necessitates the tokamak plasma in a high confinement state known as H-mode. A major challenge encountered in the standard H-mode is the abrupt collapse of its ultrathin edge layer. Here, we report a first experimental demonstration that a stable H-mode can be sustained by a staircase plasma profile. The basic physical picture is: by staircasing the plasma profile, the potential free energy stored in one single large edge layer is distributed into multiple mesoscale layers throughout the entire plasma volume. Consequently, the large scale edge collapse is converted into small scale desynchronized fluctuations, effectively reducing the risk of violent relaxation event. This finding uncovers a new benign high confinement regime of the tokamak plasma.

  • Open Access

    Augmenting the extrapolation capability of disruption prediction to extended parameter regimes by predict-first neural network

    by Zongyu Yang, Wulyu Zhong, Jiyuan Li, Yihang Chen, Da Li, Junzhao Zhang, Xia Fan, B Li, Yipo Zhang, Y B Dong, et al.

    The prediction, mitigation, and avoidance of disruptions are critical prerequisites for ensuring the safe operation of future fusion reactors. While machine learning-based disruption prediction techniques have achieved high accuracy in recent years, their application to a new device demands strong extrapolation capability across parameter regimes. This capability is essential to maintain robust algorithm performance as the operational space of the device expands, while it is also a main shortcoming for most machine learning techniques. This study evaluates the extrapolation capability of a disruption prediction algorithm using multiple subsets of experimental data from HL-3 tokamak, divided by key plasma parameters, yielding three key findings. Firstly, the area under the receiver-operator characteristic curve (AUC) of a standard deep learning algorithm trained on the subset with low parameters degrades to around 0.750, 0.666, and 0.540 when extrapolated to three progressively distant extrapolation regimes. However, the Predict-First Neural Network (PFNN) maintains an AUC of about 0.873, 0.823 and 0.701. Secondly, the algorithm's accuracy shows a clear contrast in degradation behavior: when extrapolating along electromagnetic parameters like plasma current, performance declines slowly and continuously; when extrapolating along energy-related parameters like stored energy, performance remains initially stable but deteriorates abruptly beyond a certain threshold. Finally, after rigorously accounting for the impact of parameter regime extrapolation and adjusting the algorithm deployment strategy, deep learning proved effective in protecting the device during HL-3's high-current and high-beta commissioning phases. This study provides valuable experience for implementing disruption prediction algorithms on new devices and is expected to offer practical reference for the initial operation of future fusion facilities like ITER.

  • Open Access

    A novel high-performance confinement regime in a tokamak driven by tearing mode locking and rotation shear

    by Qinghao Yan, Da Li, Deliang Yu, Bo Li, Shuosu Yang, Rui Ke, Shaobo Gong, Yihang Chen, Junzhao Zhang, Yi Zhang, et al.

    In magnetically confined fusion plasmas, tearing mode (TM) locking is conventionally regarded as detrimental, often halting rotation and triggering disruptions. Contrary to this expectation, we demonstrate that locked TMs can enhance confinement by inducing an internal transport barrier (ITB). Specifically, the interaction between a locked outer plasma region and neutral beam injection (NBI)-driven core rotation generates a broad and intense toroidal velocity shear layer, which suppresses turbulence and improves core confinement. Beam emission spectroscopy and charge exchange recombination spectroscopy confirm a substantial reduction in turbulent transport and a rise in core ion temperature exceeding T i > 9 keV . This phenomenon—termed the Locking-Induced Flow and Transport (LIFT) mode—transforms a traditionally disruptive mechanism into a confinement enhancement strategy.

  • Open Access

    Radial propagation of ELM filaments and boundary turbulence characteristics with divertor heat flux mitigation in the small-ELM regime on HL-2A

    by Ting Wu, Lin Nie, Zhanhui Wang, Jinming Gao, Yukun Shu, Yi Zhang, Yu Zhou, Yujie Zhou, Yiren Zhu, Yihang Chen, et al.

    This study investigated the radial propagation of edge-localized-mode (ELM) filaments and boundary turbulence characteristics with divertor heat flux mitigation in the small-ELM regime on the HL-2A tokamak. Divertor nitrogen seeding mitigated divertor heat flux while maintaining high plasma performance ( H 98 > 1.0 and β N > 2.0 ) , with negligible degradation of total stored energy and energy confinement time. In the pedestal, the pressure gradient and ELM amplitudes decreased after divertor heat flux mitigation. Magnetic perturbation diminished while electrostatic turbulence increased, with poloidal velocity shear remaining unchanged in the pedestal. Turbulent particle flux and perpendicular heat flux decreased in the far scrape-off layer (SOL) region. The ELM-filament burst rates and conditionally averaged amplitudes from the pedestal to the near-SOL and far-SOL regions were systematically investigated before and after heat flux mitigation. In the pedestal, the ELM-filament burst rates increased, whereas the ELM-filament amplitudes decreased after heat flux mitigation. The evolution of near-SOL filaments was partially reflected by that at the outer divertor target ( ρ ≈ 0.99 − 1.09 ). At the divertor target, the filament burst rates were larger at ρ ≈ 1.01 but became smaller at ρ ≈ 1.09 . The burst rate of filaments in the far-SOL region ( ρ ≈ 1.16 ) decreased dramatically, resulting in a significant reduction in filament transport after divertor heat flux mitigation. These results provide deeper insights into the radial propagation of ELM filaments and aid in understanding power deposition on the first wall in the small-ELM regime.

  • Integrated framework for unstable event identification of disruption on the HL-3 tokamak

    by J R Wen, Y Yu, Z Y Yang, Y Ling, B Li, Y H Chen, D Li, W L Zhong

    For tokamaks like the HL-3, which operates at reactor-grade parameters, major disruptions under high-performance conditions are intolerable. High-parameter tokamaks require not only algorithms and engineering solutions for disruption mitigation but also systematic analysis of disruption causes. Traditional disruption analysis relies on various diagnostic data, requiring both temporal evolution analysis and diagnostic cross-channel comparison, which demands substantial expert knowledge and manual effort. To address these limitations, a Plasma Event Identification System have been developed to detect key plasma events in the HL-3 tokamak. This system integrates artificial intelligence, threshold-based judgments to identify disruptions and pre-disruption plasma events including vertical displacement event, magnetohydrodynamics instability, low- q disruption and high density. The performance of each module in the system was evaluated: the disruption identification module achieved 99% accuracy; The disruption time identification module reached 95% accuracy; The tearing mode and locked mode identification attained 95% overall accuracy, demonstrating the system’s robust performance. The system enables rapid statistics and analysis for HL-3 experiments, including disruption cause analysis based on expert-logic judgments and 200 ms pre-disruption event-chain. Furthermore, this system has been deployed in the HL-3 plasma display for disruption analysis. The analysis results are also upload to the Dig Data analysis database to support the research needs of engineering and physics researchers.

  • Open Access

    HL-3 research towards high-performance plasma and power exhaust solution

    by Wulyu Zhong, Xiao Quan Ji, Wei Chen, Xingyu Bai, Alain Becoulet, Jerome B Bucalossi, Zhe Chen, Shaoyong Chen, Zhipeng Chen, Yihang Chen, et al.

    The HL-3 tokamak program addresses critical challenges in developing integrated high-performance scenarios compatible with power exhaust demands for ITER and future reactors. Through systematic facility enhancements including auxiliary heating upgrades to 19.5 MW and AI-enabled control systems achieving 95.5% disruption avoidance, HL-3 finished a new round of exploration and validation for high-performance operation and power exhaust solution. Experiments at mega-ampere plasma currents demonstrated a reactor-relevant hot-ion regime with core ion temperatures exceeding 10 keV and a fusion triple product of . High normalized beta scenarios featuring internal and double transport barriers were successfully established. To address the compatibility of the plasma boundary, various small- or no-edge localized mode (ELM) regimes were achieved, including the enhanced D-alpha (EDA) H-mode, quiescent H-mode (QH-mode), and the quasi continuous exhaust (QCE) regime. Investigations into power exhaust integration highlighted the efficacy of advanced divertor configurations, specifically snowflake and tripod geometries, in significantly reducing peak heat fluxes. Active ELM control was demonstrated via resonant magnetic perturbations (RMPs), lower hybrid waves (LHW), and impurity supersonic molecular beam injection (SMBI), complemented by real-time feedback control of divertor detachment. These developments establish the fundamental physics basis and technical foundations necessary for demonstrating and extrapolating high-performance operations to ITER and next-step devices.


  • Conversion of PF coil control voltage based on RZIP prediction model on HL-3

    by B Yang, Y H Chen, X B Zhu, M W He, X L Xu, J Sun

    For enhancing plasma parameters, stable plasma operation is a critical aspect. Precise control of the poloidal field (PF) coils promotes the exploration of plasma shape and significantly mitigates the damage caused by plasma disruptions to the device. In actual tokamak control, the adjustment of feedback control parameters and the development of new control algorithms are complex and time-consuming tasks that require substantial funding for validation and optimization. Since conventional simulation environments, though highly accurate, involve long computation times, there is an urgent need to construct a reliable RZIP (the displacement in the R direction, the displacement in the Z direction, and plasma current) response model for real-time control. This study utilizes a total of 1100 shots of historical data to develop a deep learning-based RZIP prediction model aimed at predicting RZIP data 10ms in advance. By employing a multi-task learning (MTL) approach, the model ’ s performance is enhanced to obtain a more accurate RZIP prediction model. Furthermore, by combining PID feedback control with the M -matrix (i.e. the decoupling matrix between RZIP and PF coils), a coil voltage conversion model is constructed, with RZIP as the input and PF coil control voltage as output. This research also derives more suitable PID control parameters for the current IP based on the IP data 10 ms later, leading to more reasonable PF coil control voltage data. The study predicts the RZIP using the RZIP MTL model, and the mean squared error between the predicted RZIP and the actual RZIP experimental data meets the experimental requirements. Additionally, the output voltage control for the PF coils, generated by combining PID feedback control with the M -matrix, satisfies the experimental needs.

  • Open Access

    A Heuristic Exploration of Zonal Flow-like Structures in the Presence of Toroidal Rotation in a Non-Inertial Frame

    by Xinliang Xu, Yihang Chen, Yulin Zhou, Zhanhui Wang, Xueke Wu, Bo Li, Jiang Sun, Junzhao Zhang, Da Li

    The mechanisms by which rotation influences zonal flows (ZFs) in plasma are incompletely understood, presenting a significant challenge in the study of plasma dynamics. This research addresses this gap by investigating the role of non-inertial effects—specifically centrifugal and Coriolis forces—on Geodesic Acoustic Modes (GAMs) and ZFs in rotating tokamak plasmas. While previous studies have linked centrifugal convection to plasma toroidal rotation, they often overlook the Coriolis effects or inconsistently incorporate non-inertial terms into magneto-hydrodynamic (MHD) equations. In this work, we derive self-consistent drift-ordered two-fluid equations from the collisional Vlasov equation in a non-inertial frame, and we modify the Hermes cold ion code to simulate the impact of rotation on GAMs and ZFs. Our simulations reveal that toroidal rotation enhances ZF amplitude and GAM frequency, with Coriolis convection playing a critical role in GAM propagation and the global structure of ZFs. Analysis of simulation outcomes indicates that centrifugal drift drives parallel velocity growth, while Coriolis drift facilitates radial propagation of GAMs. This work may provide valuable insights into momentum transport and flow shear dynamics in tokamaks, with implications for turbulence suppression and confinement optimization.

  • Open Access

    Model-Based Optimization of the Field-Null Configuration for Robust Plasma Breakdown on the HL-3 Tokamak

    by Muwen He, Bin Yang, Yihang Chen, Xinliang Xu, Xiaobo Zhu, Jiaqi Yang, Jiang Sun, Panle Liu, Bo Li, Xiaoquan Ji

    This paper introduces a self-consistent field-null optimization algorithm of a poloidal magnetic field that precisely accounts for the influence of vacuum vessel eddy currents. Building on existing poloidal field (PF) coil currents, the algorithm can refine these waveforms to achieve various target field-null configurations. Firstly, based on the TokSys toolbox, a response model, including the PF coils and vacuum vessel circuits for the HL-3 tokamak, is developed under the MATLAB® and Simulink™ framework. The resistivity parameters of the model are calibrated using experimental data obtained from single-coil discharge tests. Subsequently, an iterative method was employed to simultaneously solve the dynamic field-null optimization problem within a specified spatial region and precisely account for the effect of passive eddy currents. Typically, B⊥≤1 G within a large area can be obtained with this iterative scheme, which can be stably sustained for over 15 milliseconds to ensure the robustness of breakdown. Finally, a low-pass filtered PID controller is applied to the model to achieve precise control of the PF coils currents, confirming the feasibility of implementing the proposed algorithm in real experiments.