Chenguang Wan

Chenguang Wan

Auto-generated profile, based on publicly available information

Chenguang Wan

School of Physical and Mathematical Sciences, Nanyang Technological University, Singapore 637371, Singapore

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 April 2023 in Nuclear Fusion
Verified
Journal Article
Open access
PDF file
CWChenguang Wan
Chenguang Wan
ZYZhi Yu
Zhi Yu
APAlessandro Pau
Alessandro Pau
OSOlivier Sauter
Olivier Sauter
XLXiaojuan Liu
Xiaojuan Liu
JLJiangang Li
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.

29 April 2021 in Nuclear Fusion
Verified
Journal Article
Closed access
PDF file
CWChenguang Wan
Chenguang Wan
ZYZhi Yu
Zhi Yu
FWFeng Wang
Feng Wang
XLXiaojuan Liu
Xiaojuan Liu
JLJiangang Li
Jiangang Li
A neural network model of tokamak discharge is developed based on the experimental dataset of a superconducting long-pulse tokamak (EAST) campaign 2016–2018. The purpose is to reproduce the response of diagnostic signals to actuator signals without introducing additional physical models. In the present work, the discharge curves of electron densityne, stored energyWmhd, and loop voltageVloopwere reproduced from a series of actuator signals. ForneandWmhd, the average similarity between the modeling results and the experimental data achieve 89% and 97%, respectively. The promising results demonstrate that the data-driven methodology provides an alternative to the physical-driven methodology for tokamak discharge modeling. The method presented in the manuscript has the potential of being used for validating the tokamak’s experimental proposals, which could advance and optimize experimental planning and validation.

27 July 2022 in IEEE Transactions on Power Electronics
Verified
Journal Article
Closed access
XDXi Deng
Xi Deng
CWChenguang Wan
Chenguang Wan
LJLi Jiang
Li Jiang
Ge GaoGe Gao
Ge Gao
YHYa Huang
Ya Huang
The three-phase two-level pulsewidth modulation (PWM) converter operates in rectification state or active inverter state according to the corresponding experimental requirements of Experimental Advanced Superconducting Tokamak. However, the current fault diagnosis methods proposed for PWM converters usually deal with only the rectification state or inverter state. To solve this problem, this article proposes a diagnosis method to detect single and double open-circuit faults under both the rectification state and active inverter state. First, an inception-based neural network diagnosis model is proposed, in which 1 × 1, 3 × 3, 5 × 5, and 7 × 7 convolution and 3 × 3 max-pooling processes are conducted in parallel at the input and the output and stacked together to generate the final output. Second, the model is trained using three-phase alternating current current signals to determine the location of open-switch faults with a large number of simulation data samples and fewer experimental data samples. Finally, the experimental result shows that the proposed model can accurately detect approximately 99.14% of the open-switch faults within 12.83 ms (<3/4 cycle) without an additional sensor. Furthermore, the result shows that the proposed model is robust to load voltage transient conditions and nonunity power factor conditions.

10 November 2022 in Nuclear Fusion
Verified
Journal Article
Open access
CWChenguang Wan
Chenguang Wan
ZYZhi Yu
Zhi Yu
APAlessandro Pau
Alessandro Pau
XLXiaojuan Liu
Xiaojuan Liu
JLJiangang Li
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.

10 January 2024 in Nuclear Fusion
Verified
Journal Article
Open access
PDF file
CWChenguang Wan
Chenguang Wan
SBShuhang Bai
Shuhang Bai
ZYZhi Yu
Zhi Yu
QYQiping Yuan
Qiping Yuan
YHYao Huang
Yao Huang
JLJiangang Li
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.

7 April 2025 in Nuclear Fusion
Verified
Journal Article
Open access
CWChenguang Wan
Chenguang Wan
YCYoungwoo Cho
Youngwoo Cho
ZQZhisong Qu
Zhisong Qu
YCYann Camenen
Yann Camenen
RVRobin Varennes
Robin Varennes
XGXavier Garbet
Xavier Garbet
One of the main challenges in building high-fidelity surrogate models of tokamak turbulence is the substantial demand for high-quality data. Typically, producing high-quality data involves simulating complex physical processes, which requires extensive computing resources. In this work, we propose a fine tuning-based approach to develop the surrogate model that reduces the amount of high-quality data required by 80%. We demonstrate the effectiveness of this approach by constructing a proof-of-principle ion temperature gradient surrogate model using datasets generated from two gyrokinetic codes, GKW and GX. GX needs in terms of computing resources are much lighter than GKW. Remarkably, the surrogate models’ performance remain nearly the same whether trained on 798 GKW results alone or 159 GKW results plus an additional 11979 GX results. These encouraging outcomes indicate that fine tuning methods can significantly decrease the high-quality data needed to develop the simulation-driven surrogate model. Moreover, the approach presented here has the potential to facilitate surrogate model development for heavy codes and may ultimately pave the way for digital twin systems of tokamaks.