ZYZhi Yu

Zhi Yu

*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


Zhi Yu has a total of 33 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A data-driven model for stable long-horizon autoregressive prediction of plasma current and control-oriented boundary evolution in EAST

    by Minglong Wang, Chenguang Wan, Yuehang Wang, Jia Huang, Zhi Yu, Jingjing Lu, Xiaojuan Liu, Zhisong Qu, Weidong Chen, Teng Wang, et al.

    Accurate long-horizon prediction of tokamak plasma current, position, and boundary evolution is essential for magnetic control, rapid controller development, and reinforcement-learning-based optimization. However, high-fidelity physics-based simulators are often computationally prohibitive when large numbers of simulations are required for controller tuning and large-scale optimization. In this work, we develop a fast, data-driven model for the Experimental Advanced Superconducting Tokamak (EAST) to predict plasma current, position, and shape evolution over horizons of up to 1 s, with an inference time of 0.1 s on a single NVIDIA H800 GPU. Evaluated on a temporally separated test set comprising 5907 discharges, the model demonstrates strong generalization to evolving operating conditions and maintains stable agreement with experimental measurements during long-horizon autoregressive rollouts. The mean geometric error of the plasma centroid and the control-oriented boundary representation is 2.4 cm, indicating reliable centimeter-level accuracy for control-oriented prediction of plasma position and shape evolution. The proposed framework provides a practical, high-throughput alternative to first-principles simulators, enabling efficient control algorithm prototyping, large-scale scenario exploration, and future data-driven optimization for tokamak plasma control. This study focuses on the post-shaping phase of EAST discharges, where sustained plasma current and control-oriented boundary regulation are most critical.

  • Open Access

    Bayesian neural networks for predicting tokamak energy confinement time with uncertainty quantification

    by Enliang Gao, Chenguang Wan, Youjun Hu, Minglong Wang, Jingjing Lu, Zhisong Qu, Xinghao Wen, Jia Huang, Ying Chen, Heru Guo, et al.

    Accurate estimation of the tokamak energy confinement time (τ E ) is crucial for optimizing the operation and design of fusion devices. Traditional methods, such as the ITER scaling law, often lack predictive precision and provide limited uncertainty quantification. To address these limitations, we develop two Bayesian neural network (BNN) models: VIBNN and NUTSBNN. These models leverage variational inference (VI) and the No-U-Turn Sampler (NUTS), respectively, for posterior inference. Evaluations on the multi-machine ITPA global H-mode confinement database and individual datasets from JET, DIII-D, and ASDEX-Upgrade, both models outperform conventional methods in terms of accuracy and provide reliable uncertainty estimates. NUTSBNN achieves higher accuracy with an inference time of ∼3 s, making it suitable for offline analysis, while VIBNN provides faster inference ( ∼ 0.05 s), which is promising for real-time applications. These results highlight the potential of BNNs for accurate and interpretable confinement time predictions in fusion research.


  • Open Access

    Time series extrinsic regression for reconstructing missing electron temperature in tokamak

    by Minglong Wang, Chenguang Wan, Jingjing Lu, Zhi Yu, Bingjia Xiao, Yanlong Li, Xiaoxue He, Zhengping Luo, Qiping Yuan, Yemin Hu, et al.

    The integrity of tokamak plasma diagnostic data is critical for physics research and the development of experimental fusion device. Sensor failures, data acquisition errors, or limitations in diagnostic systems (e.g. absent electron temperature diagnostic) pose significant challenges for experimental data analysis, physical integrated simulation, and the design and optimization of tokamak experiment. To address this, we propose a data-driven machine learning approach based on time series extrinsic regression (TSER) to reconstruct missing electron temperature data in tokamak experiments. By constructing and training TSER models, we can effectively reconstruct missing electron temperature measurement while capturing the complex interrelationships within multiple time series signals. This study provides a solution for enhancing the integrity and reliability of tokamak diagnostic, thereby strengthening the foundation for future fusion research. Comprehensive test and evaluation results show that our method is capable of achieving a confidence level of over 95.9% to ensure that the reconstruction of electron temperature throughout the discharge process is within double the standard deviation error of the true value, meeting the stringent requirements of physics research and integrated simulations.


  • Open Access

    Predict the last closed-flux surface evolution without physical simulation

    by Chenguang Wan, Shuhang Bai, Zhi Yu, Qiping Yuan, Yao Huang, Xiaojuan Liu, Yemin Hu, Jiangang Li

    One of the main challenges in developing effective control strategies for the magnetic control system in tokamaks has been the difficulty in obtaining the last closed-flux surface (LCFS) evolution results from control commands. We have developed a data-driven model that combines a predictive model and a surrogate model for physics simulation programs. This model is capable of predicting the LCFS without relying on physical simulation codes. Addressing the data characteristics of LCFS, we have proposed a specialized discretization approach to achieve dimensionality reduction. Furthermore, we have excluding the control references, the model can be seamlessly integrated into the control system, providing real-time LCFS prediction. Following comprehensive testing and multifaceted evaluation, our model has demonstrated highly satisfactory results of 95% or above, meeting practical requirements.


  • Open Access

    Applying FAIR4RS principles to develop an integrated modeling environment for the magnetic confinement fusion

    by Xiaojuan Liu, Zhi Yu, Nong Xiang

    Over the decades, the integrated modeling (IM) environment for magnetically confined fusion has evolved from a single, isolated, proprietary numerical computing software to an open, flexible platform emphasizing sharing, communication, and workflow. This development direction is consistent with the FAIR4RS principles put forward by the scientific community in recent years. In this article, we describe how the FAIR4RS principles were put into practice during the development of the IM management tool FyDev for the Experimental Advanced Superconducting Tokamak (EAST). FyDev integrates the process of building, deploying, and invoking research software, automating the entire process. FyDev can also assign a unique ID for each software, convert the software ID to a Python module, and encapsulate a package management tool to enhance the software building process, ensuring consistency throughout the entire phase of the research software find, access, use, and invocation in a uniform contextual environment.


  • Open Access

    Corrigendum: EAST discharge prediction without integrating simulation results (2022 Nucl. Fusion 62 126060)

    by Chenguang Wan, Zhi Yu, Alessandro Pau, Xiaojuan Liu, Jiangang Li

    In the published paper titled ‘EAST discharge prediction without integrating simulation results’ (2022 Nucl. Fusion 62 126060), figure 5 used the wrong plot, and is not matched with table 3. We used the plot discussed with referees, the case when we do not use IC 1 as a proxy. This corrigendum provides the correct plot for figure 5.


  • Open Access

    A machine-learning-based tool for last closed-flux surface reconstruction on tokamaks

    by Chenguang Wan, Zhi Yu, Alessandro Pau, Olivier Sauter, Xiaojuan Liu, Qiping Yuan, Jiangang Li

    Tokamaks allow to confine fusion plasma with magnetic fields. The prediction/reconstruction of the last closed-flux surface (LCFS) is one of the primary challenges in the control of the magnetic configuration. The evolution in time of the LCFS is determined by the interaction between the actuator coils and the internal tokamak plasma. This task requires real-time capable tools to deal with high-dimensional data and high resolution at same time, where the interaction between a wide range of input actuator coils with internal plasma state responses adds an additional layer of complexity. In this work, we present the application of a novel state-of-the-art machine learning model to LCFS reconstruction in an experimental advanced superconducting tokamak (EAST) that learns automatically from the experimental data of EAST. This architecture allows not only offline simulation and testing of a particular control strategy but can also be embedded in a real-time control system for online magnetic equilibrium reconstruction and prediction. In real-time modeling tests, our approach achieves very high accuracies, with an average similarity of over 99% in the LCFS reconstruction of the entire discharge process.


  • Open Access

    EAST discharge prediction without integrating simulation results

    by Chenguang Wan, Zhi Yu, Alessandro Pau, Xiaojuan Liu, Jiangang Li

    In this work, a purely data-driven discharge prediction model was developed and tested without integrating any data or results from simulations. The model was developed based on the experimental data from the Experimental Advanced Superconducting Tokamak (EAST) campaign 2010–2020 discharges and can predict the actual plasma current I p, normalized beta β n, toroidal beta β t, beta poloidal β p, electron density n e, stored energy W mhd, loop voltage V loop, elongation at plasma boundary κ, internal inductance l i, q at magnetic axis q 0, and q at 95% flux surface q 95. The average similarities of all the selected key diagnostic signals between prediction results and the experimental data are greater than 90%, except for the V loop and q 0. Before a tokamak experiment, the values of actuator signals are set in the discharge proposal stage, with the model allowing to check the consistency of expected diagnostic signals. The model can give the estimated values of the diagnostic signals to check the reasonableness of the tokamak experimental proposal.