WLWenbin Liu

Wenbin Liu

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Wenbin Liu has a total of 174 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Cross‐device and cross‐modal arc detection: transfer learning between EAST visible and WEST infrared diagnostics

    by Zhongfang Guan, Erwan Grelier, Bin Zhang, Sebastien Vives, Baoguo Wang, Victor Moncada, Léo Dubus, Jinping Qian, Kedong Li, Jian Liu, et al.

    This work presents an engineering-oriented data processing and model training pipeline for Experimental Advanced Superconducting Tokamak (EAST) visible and (W) Environment in Steady-state Tokamak (WEST) infrared data, enabling a systematic study of cross-device and cross-modal electric arc detection. A lightweight convolutional neural network is adopted as a cross-device arc detection model, and joint training and transfer learning experiments are conducted on EAST visible (VIS) and WEST infrared (IR) data. Single‐modality baselines are first trained on EAST and WEST, then evaluated zero‐shot on the opposite device, followed by a comparison of frozen‐backbone and fine‐tuned transfer for VIS → IR and IR → VIS. The results show that models trained from scratch on each diagnostic channel already achieve high precision and recall for arc detection, whereas direct zero-shot cross-modal transfer drives the arc recall on the opposite domain almost to zero. In contrast, starting from the single-modality baselines and modestly fine tuning higher convolutional layers and the classification head on the target domain can significantly restore-cross device, cross-modal detection performance, with positive class precision and recall approaching or even surpassing the single-domain baselines. Class activation map analysis further indicates that, under fine-tuned and joint training configurations, the models consistently focus on the antenna, the arc, and its surrounding scattering region, supporting physically meaningful explanations and suggesting that cross-device, cross-modal training can capture arc related structures that generalize across devices while providing a lightweight, scalable route to reusing imaging-based diagnostic knowledge with reduced annotation and retuning effort. Given that machine protection is safety-critical and that imaging diagnostics are never identical across devices, the transferability demonstrated in this work is directly relevant to equipping next-generation tokamaks such as ITER and BEST with a reliable arc-detection capability before a large native training set becomes available. On a newly commissioned device, arc events are rare, sporadic and safety-relevant, so deferring detection until a sufficiently large native training corpus has accumulated would leave the machine unprotected during the early operational phase, when limited operating experience makes the associated risks most difficult to characterize; cross-machine and cross-modal transfer instead enables a new device to inherit a usable detection capability from the outset. Accordingly, the present work is intended not merely as a benchmark of cross-domain accuracy, but as evidence that scalable, transferable arc monitoring can constitute a machine-protection layer of future fusion reactors.

  • Open Access

    Progress of understanding the effect of helium concentration on ion–electron collision and energy confinement on EAST

    by Qian Wang, Bin Zhang, Xianzu Gong, Jinping Qian, Xiaoming Zhong, Kedong Li, Wenbin Liu, Tianqi Jia, Yutong Guo, Wei Wang, et al.

    Helium plasma experiments conducted on different tokamaks all demonstrate that the energy confinement are lower than those of deuterium plasmas under similar operation conditions. However, the origin of the difference in confinement between helium and hydrogen isotopes remains unclear. To understand the influence of helium concentration, which leads to the variation of ion mass and charge number, on energy confinement, new helium experiments were conducted in the EAST tokamak with dominant electron heating and a tungsten divertor in 2025. A notable dependence on both ion mass and charge number is observed in the effective diffusion coefficient, consistent with the trend predicted by the gyro-Bohm scaling χ gB     ∝ ρ i     . The higher energy confinement time is observed with a lower ion mass. Conversely, when both the ion mass and charge number are increased, the energy confinement time is found to be similar. Power balance and transport analyses indicate that the ion-electron collision, significantly influenced by the variation of ion mass and charge number, appears to play a dominant role in regulating ion-scale turbulence in helium H-mode discharges. Furthermore, linear electromagnetic gyrokinetic simulation reveals that the linear growth rates of electron temperature gradient (ETG) modes might be reduced by the enhancement of ion-electron energy exchange in L-mode discharges. Consequently, it is crucial to take into account the role of electron-ion collision/energy exchange to understand the influence of ion mass and chargenumber on energy confinement. These results contribute to a better understanding of the transport characteristics in multi-ion-component plasma.

  • Open Access

    Tokamak discharge simulation for EAST by coupling METIS and free-boundary equilibrium code FBT

    by Wenyi Lu, Wenbin Liu, Jinping Qian, Miaohui Li, Ye Tao, Zhengping Luo, Dehong Chen, Rundu Hu, Feifei Long, Jian Liu, et al.

    Fast and reliable discharge modeling is an essential tool for tokamak scenario development, as extensive simulations are required to explore the feasible operational space and translate physical targets into implementable actuator waveforms. In this work, a fast discharge simulator for EAST was developed by coupling the fast integrated tokamak modeling tool METIS with the free-boundary equilibrium code FBT from the MEQ (Matlab EQuilibrium) suite. METIS provides the evolution of the plasma profiles by self-consistently solving heat and particle transport with source profiles. The generated outputs are then used directly in FBT to computes the poloidal field coil currents needed in order to obtain a given plasma shape. The coupling strategy is based either on matching the free functions (p'(ψ) and FF'(ψ)) on the right-hand side of the Grad–Shafranov equation, or on matching global quantities such as the plasma current Ip and stored energy W_MHD. Both approaches are effective in ensuring consistency of the pressure and current-density profiles between the two codes. The last closed flux surface (LCFS) curve computed by FBT is fed back into METIS to update the LCFS, and METIS is then run again. The METIS–FBT iteration is repeated until convergence is reached. The results obtained with the equilibrium code FBT are first benchmarked against experimental data and reconstruction results available on EAST. The capability of the coupled codes is demonstrated by simulating a radio-frequency heated discharge on EAST, showing that the integrated discharge simulator enables scenario prediction that is both physically consistent and engineering-feasible. Finally, the application of this new simulation tool for tokamak scenario design is discussed.

  • Open Access

    Accessing the density-free regime with ECRH-assisted ohmic start-up on EAST

    by Jiaxing Liu, Ping Zhu, Dominique Franck Escande, Wenbin Liu, Shiwei Xue, Xin Lin, Panjun Tang, Liang Wang, Ning Yan, Jinju Yang, et al.

    High plasma density operation is crucial for a tokamak to achieve energy breakeven and burning plasma. However, there is often an empirical upper limit of electron density in tokamak operation, namely, the Greenwald density limit n G , above which tokamaks generally disrupt. Achieving high-density operation above the density limit has been a long-standing challenge in magnetic confinement fusion research. Here, we report experimental results on the Experimental Advanced Superconducting Tokamak (EAST) achieving line-averaged electron density in the range of (1.3 to 1.65) n G , significantly above the typical EAST operational range of (0.8 to 1.0) n G . This is performed with electron cyclotron resonance heating (ECRH)–assisted ohmic start-up and sufficiently high initial neutral density. These experiments are shown to operate in the density-free regime first predicted by a recent plasma-wall self-organization theory. These results suggest a promising scheme for substantially increasing the density limit in tokamaks, a critical advancement toward achieving burning plasma.


  • Open Access

    Validation of DYON simulations and development of physical sputtering models for lithiation and boronisation in EAST

    by Runze Chen, Hyun-Tae Kim, Wenbin Liu, Jinping Qian, Bin Zhang, Ximan Li, Zhengping Luo, Dalong Chen, Yaowei Yu, Wei Gao, et al.

    This paper validates for the first time the predictive capability of the DYON code for plasma initiation in EAST, which has metallic wall, superconducting coils and conventional tokamak shape, like ITER. The model accurately reproduced the operating spaces of loop voltage and prefill gas pressure for ohmic discharges, demonstrating its validity in predicting the required operating parameters for successful inductive plasma initiation in EAST. The role of wall conditioning on plasma initiation was investigated with the newly developed physical sputtering models of Boron and Lithium. In EAST experiments, it was observed that the discharges after boronisation of the wall are much more vulnerable to plasma burn-through failure than after lithium-coating. The simulation results revealed that despite the similar physical sputtering yield in Boron and Lithium, the radiative energy loss rates for the boron-coated wall are significantly higher than those for the lithium-coated wall, due to the much higher radiative power coefficients of Boron. Parametric scans of initial Boron content in ohmic discharge at the typical prefilled gas pressure in EAST (0.8 mPa) showed that even 1.5% of Boron content in the prefilled gas, possibly remaining after boronisation of the wall, could lead to excessive radiation energy losses and failure of plasma burn-through. For successful plasma burn-through with 1.5% initial boron content, the modelling indicates 10 kW absorption of EC power is required, and it increases with more initial boron e.g. 50 kW for 3% initial boron content.


  • Open Access

    Thermal damage detection of EAST internal component based on machine learning

    by Zhongfang Guan, Bin Zhang, Jian Liu, Jinping Qian, Xianzu Gong, Runze Chen, Zuhao Wang, Binfu Gao, Yutong Guo, Chuannan Xuan, et al.

    During the discharge process, thermal events are common and destructive in Tokamak experiments. High energy particles in plasma can collide with device components such as limiters and diverters, which can lead to overheating, material cracking, and even damage to the structure of the device. Therefore, we need efficient detection methods to monitor thermal events in real-time. To address the challenge of identifying thermal damage to internal components of the first wall during the Experimental Advanced Superconducting Tokamak (EAST) discharges, we introduce the YOLOv8 model for hotspot detection on EAST and present the customized EAST- You Only Look Once (YOLO) algorithm, derived from an enhanced YOLOv8 framework. YOLOv8, known for its strong performance in real-time object detection, serves as a robust base model for this task. However, its performance on small object detection, such as early-stage thermal damage, is limited. Improvements are needed for specialized tasks, particularly for early warning and precise identification of small internal component damage during the initial stages of EAST discharges. we enhance the YOLOv8 model by incorporating specialized layers for detecting small targets and integrating the CBAM attention mechanism. These adjustments result in a network model capable of sensitively detecting internal component damage in EAST. Experimental results demonstrate that EAST-YOLO surpasses several versions of traditional YOLOv8 models in model evaluation metrics, achieving a precision of 97.5%, mAP50 of 97.7%, and a Recall of 94.0%. This problem- oriented approach significantly improves the operational safety and stability of the EAST device by enabling early detection of thermal events. The AI-based detection method provides a new solution to safeguarding the fusion device, while also offering potential avenues for integrating artificial intelligence technologies into EAST feedback control and operating systems in the future.


  • Open Access

    Development of plasma burn-through simulation code and validation in SUNIST-2 and EAST

    by Ximan Li, Zichong Song, Yunxiao Wei, Runze Chen, Wenbin Liu, Zhengbo Cheng, Jinping Qian, Yi Tan, Long Zeng, Zhe Gao

    A full electromagnetic plasma burn-through simulation code has been developed to improve understanding of the plasma breakdown and initiation process in tokamaks, including spherical tokamaks. A 0D plasma model is adopted in the model, with only operation signals as input, such as current waveforms of central solenoid coils and poloidal field coils, prefill gas pressure and parameters for wall conditioning, which are determined before discharge experiments. The output includes the time evolution of the plasma current, electron temperature, ion temperature, and densities of gas compositions. The simulation code mainly contains three parts: circuit equations, energy balances, and particle balances. Eddy currents on toroidal passive conducting units of vacuum vessel and plasma current are calculated simultaneously by solving circuit equations. The use of eddy current waveforms and coil current waveforms enables calculation of the 2D space distribution of time-evolving poloidal magnetic field and flux, and consequently the plasma volume evolution can be calculated using the last closed flux surface. Energy balances and particle balances are used to solve the temperature and density evolution of fuel gas and impurities. The main procedures in energy balances and particle balances are ionization, recombination, charge exchange, radiation, and transport. Transport loss is estimated by perpendicular loss by Bohm diffusion and parallel loss along field lines, which is calculated using connection length and ion sound speed. The connection length is calculated using magnetic field line tracing with electromagnetic modeling. The code has been used to reproduce the time evolution of plasma current, electron density, flux loop data, and H α or D α data of plasma burn-through on SUNIST-2 spherical tokamak and EAST tokamak, and the simulated results are consistent with experimental measurements with the relative errors of the plasma current, loop voltage, and magnetic flux under 15%.


  • Research on confinement improvement under optimized injection time of electron cyclotron resonance heating power at EAST

    by Z H Wang, B Zhang, X Z Gong, J P Qian, W Wang, Y T Guo, W B Liu, P Li, Q L Wang, Y C Hu, et al.

    A primary objective of the experimental advanced super-conducting tokamak (EAST) is to demonstrate steady-state long-pulse high-performance plasma operation for future large-scale devices like ITER and CFETR (Wan et al 2017 Nucl. Fusion 57 102019). The formation of internal transport barriers (ITBs) is one of the key issues to achieve high-performance plasma operation. Optimizing the current density profile is a promising way to improve plasma confinement, which is beneficial to the formation of ITBs. A lot of effort has been dedicated to optimizing the current density profile at EAST over recent years. In this paper, the authors discuss the formation of ITBs leading to a significant improvement of plasma confinement by optimizing the injection time of electron cyclotron resonance heating (ECRH) power at EAST. After ECRH delayed injection, poloidal beta and thermal energy confinement ( H 98 y 2 ) was observed to increase. The current density distribution and the power deposition distribution were changed. The analysis of the soft x-ray (Xu et al 2020 Phys. Scr. 95 055603) and electron cyclotron emission (Liu et al 2016 Plasma Sci. Technol. 18 1148–54) diagnostics data showed that the structural strength of magnetohydrodynamics was significantly decreased in the core region in this case. The turbulence growth rate calculated by the tokamak global linearized fusion code (Kinsey et al 2008 Phys. Plasmas 15 055908) also shows that trapped electron mode turbulence is stabilized.


  • Open Access

    ECW assisted plasma startup with low toroidal electric field and full metal wall in EAST superconducting tokamak

    by Wenbin Liu, Jinping Qian, Hongjuan Sun, Hyun-Tae Kim, Runze Chen, Bin Zhang, Handong Xu, Zhengping Luo, Yaowei Yu, Wei Gao, et al.

    Recent experimental results from the EAST superconducting tokamak with a full metal wall (without Li-coating) provide strong support for the ITER startup scheme with electron cyclotron wave (ECW) assistance. Detailed investigations were performed on the impact of different magnetic field geometries, including quadrupolar field configuration (QFC), wide null field configuration (NFC), and trapped particle configuration (TPC). All three configurations enabled the breakdown of neutral gas with ECW pre-ionization. However, in most discharges with QFC, the absence of closed flux surfaces after the breakdown was associated with unsuccessful plasma startups. Conversely, both NFC and TPC achieved successful plasma startups under normal conditions, with TPC demonstrating higher robustness against high impurity levels compared to NFC. Experiments found that the minimum ECW power required for a successful TPC startup is approximately 0.48 MW. Plasma startup failed if the prefill gas pressure exceeded optimal levels for a given ECW power ( ⩾ 0 .48 MW), with the critical gas pressure linearly dependent on the ECW power. Additionally, higher ECW power was required with increased impurity levels. With ECW assistance, the toroidal electric field at R = 1.85  m in EAST routine operations ( < 0.15  V m − 1 ) is significantly lower than ITER’s maximum value (0.3 V m−1), suggesting the possibility of reducing flux consumption and lowering demands on coils and power supplies for ITER.


  • Open Access

    Experimental research of ECW pre-ionization and assisted startup in EAST

    by Runze Chen, Wenbin Liu, Jinping Qian, Bin Zhang, Hongjuan Sun, Zhengping Luo, Yaowei Yu, Wei Gao, Miaohui Li, Handong Xu, et al.

    Experimental research on the electron cyclotron wave (ECW) pre-ionization and assisted start-up was carried out systematically for the first time in EAST tokamak, which is a superconducting device with ITER-like full metal wall. Breakdown and plasma initiation at low toroidal electric fields (−1) with ECW pre-ionization and startup assistance has been demonstrated. Also, the parameter domain of breakdown is significantly extended towards higher prefill gas pressure. The effect of ECW injection timing, power, toroidal injection angle on breakdown were also investigated. Injecting ECW earlier leads to an earlier breakdown and a higher plasma current ramp rate. The electron cyclotron heating (ECH) power threshold for breakdown in EAST is approximately 0.4 MW. In the range of ECH power tested in this work, higher ECH power is advantageous for achieving earlier and faster breakdown. Furthermore, the breakdown with radial ECW injection occurs earlier compared with oblique injections (co-current and counter-current). During the ECW-assisted startup, the process of burn-through is prolonged by the higher pre-filled gas pressure even though it enhances the ease of breakdown. In addition, compared to the low hybrid wave assistance, the ECW assistance has an effect in averting the generation of runaway electrons and improving the safety of device during startup. Moreover, the ECW assistance exhibits a high tolerance to the impurity and thus ensures a high ramp rate of plasma current even with a high impurity level.