Abstract:In the field of magnetic confinement plasma control, the accurate feedback of plasma position and shape primarily relies on calculations derived from magnetic measurements through equilibrium reconstruction or matrix mapping method. However, under harsh conditions like high-energy neutron radiation and elevated temperatures, the installation of magnetic probes within the device becomes challenging. Relying solely on external magnetic probes can compromise the precision of EFIT in determining the plasma shape. To tackle this issue, we introduce a real-time, non-magnetic measurement method on the HL-3 tokamak, which diagnoses the plasma position and shape via imaging. Particularly, we put forward an adapted Swin Transformer model, the Poolformer Swin Transformer (PST), to accurately and fastly interpret the plasma shape from the Charge-Coupled Device Camera (CCD) images. By adopting multi-task learning and knowledge distillation techniques, the model is capable of robustly detecting six shape parameters under disruptive conditions such as a divertor shape and gas injection, circumventing global brightness changes and cumbersome manual labeling. Specifically, the well-trained PST model capably infers R and Z within the mean average error below 1.1 cm and 1.8 cm, respectively, while requiring less than 2 ms for end-to-end feedback, an 80 improvement over the smallest Swin Transformer model, laying the foundation for real-time control. Finally, we deploy the PST model in the Plasma Control System (PCS) using TensorRT, and achieve 500 ms stable PID feedback control based on the PST-computed horizontal displacement information. In conclusion, this research opens up new avenues for the practical application of image-computing plasma shape diagnostic methods in the realm of real-time feedback control.

Real-time plasma boundary shape reconstruction using visible camera on EAST Tokamak

Optical Plasma Boundary Reconstruction Based on Least Squares for EASTTokamak

Kinetic Equilibrium Reconstruction on EAST Tokamak

System identification for plasma shape control on EAST

EAST Kinetic Equilibrium Reconstruction Combining with Polarimeter-Interferometer Internal Measurement Constraints

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

Novel identification algorithm for plasma boundary gap based on visible endoscope diagnostic on EAST tokamak

Real-Time Measurements for Edge Plasma Parameters Using Triple Probes on East

Remote in Situ Laser-Induced Breakdown Spectroscopic Approach for Diagnosis of the Plasma Facing Components on Experimental Advanced Superconducting Tokamak.

Tomographic Reconstruction Of Emissive Profile In The Divertor Region For The Visible Light Imaging Diagnostic On Experimental Advanced Superconducting Tokamak

Equilibrium reconstruction constrained by the consistency of current simulation on EAST

Development of ultra-short pulse reflectometry on the Experimental Advanced Superconducting Tokamak (EAST)

Adapted Swin Transformer-based Real-Time Plasma Shape Detection and Control in HL-3

Equilibrium Reconstruction Based on Core Magnetic Measurement and Its Applications on Equilibrium Transition in Joint-TEXT Tokamak

East-Aia Deployment under Vacuum: Calibration of Laser Diagnostic System Using Computer Vision

An Efficient Technique for Magnetic Equilibrium Reconstruction with Q Profile Constraints and Its Application on the East Tokamak

Reconstruction of Poloidal Magnetic Fluxes on EAST based on Neural Networks with Measured Signals

Techniques Used for Registration and Reconditioning of Infrared Images of the Experimental Advanced Superconducting Tokamak (EAST) Divertor

A new probe design to measure plasma properties and toroidal correlations on the EAST

Safety factor diagnostic for tokamak core plasma from three-dimensional reconstruction of pellet ablation trail