A neural network model, EFITNN, has been developed capable of real-time magnetic equilibrium reconstruction based on HL-3 tokamak magnetic measurement signals. The model processes inputs from 68 channels of magnetic measurement data gathered from 1159 HL-3 experimental discharges, including plasmas current, loop voltage, and the poloidal magnetic fields measured by probes. The outputs of the model feature eight key plasmas parameters, alongside high-resolution () reconstructions of the toroidal current density and poloidal magnetic flux distributions Ψrz. Moreover, the network's architecture employs a multi-task learning structure, which enables the sharing of weights and mutual correction among different outputs, and leads to increase the model's accuracy by up to 32%. The performance of EFITNN demonstrates remarkable consistency with the offline EFIT, achieving average 0.941, 0.997 and 0.959 for eight plasmas parameters, Ψrz and , respectively. The model's robust generalization capabilities are particularly evident in its successful predictions of quasi-snowflake (QSF) divertor configurations and its adept handling of data from shot numbers or plasmas current intervals not previously encountered during training. Compared to numerical methods, EFITNN significantly enhances computational efficiency with average computation time ranging from 0.08 ms to 0.45 ms, indicating its potential utility in real-time isoflux control and plasmas profile management.
This paper presents a neural network model, EFITNN, that can quickly and accurately reconstruct the magnetic equilibrium of a tokamak plasma in real-time using data from HL-3. The model predicts key plasma parameters and detailed current and flux distributions, outperforming traditional numerical methods in both accuracy and speed.