The EAST research program aims to demonstrate steady-state long-pulse advanced high-performance H-mode operations with ITER-like poloidal configuration and RF-dominated heating schemes. Since last IAEA FEC, EAST has been upgraded with all ITER-relevant auxiliary heating and current drive systems, enabling the investigation of plasma profile control by coupling/integration of various combinations. By means of the 4.6 GHz and 2.45 GHz LHCD systems, H-mode can be obtained and maintained at relatively high density, even up to n e ~ 4.5 × 10 19 m-3 , where a current drive effect is still observed. Significant progress has been achieved on EAST, including: i). Demonstration of a steady-state scenario (fully non-inductive with V loop ~ 0.0V at high β P ~ 1.8 and high performance (H 98,y2 > 1.0) in upper single-null (ε ~ 1.6) configuration with the tungsten divertor; ii) Discovery of a stationary ELM-stable H-mode regime with 4.6 GHz LHCD; iii) achievement of ELM suppression in slowly-rotating H-mode plasma with the application of n = 1 and 2 RMPs.

The simulations on edge-localized modes (ELMs) with six-field peeling–ballooning (P–B) modes using the BOUT++ code are reported in this paper. This six-field model based on the full Braginskii equations are developed to simulate self-consistent turbulence and transport between ELMs. Through the comparison with the previous three-field two-fluid model, P–B instability, ion diamagnetic effects, resistivity and hyper-resistivity are found to be the dominant physics during ELMs. The additional physics, such as ion acoustic waves, thermal conductivities, Hall effects, toroidal compressibility and electron–ion friction, are less important in this process. Through the simulations within different equilibrium temperature profiles but with the same pressure and current, the particle loss of ions contributes the least to the total ELM size. The ELM size will be smaller for low-density cases. The study of convective particle and heat flux indicates that the peak of radial particle flux is obviously related to the ELM filaments burst events. The analysis of radial transport coefficients indicates that the ELM size is mainly determined by the energy loss at the crash phase. The typical values for transport coefficients in the saturation phase after ELM crashes are Dr ∼ 200 m2 s−1, χir ∼ χer ∼ 40 m2 s−1. The turbulent zonal flow, which is mainly driven by the Reynolds stress and suppressed by ion diamagnetic terms, regulates the turbulence from the ELM crash phase to the quasi-steady state for large ELM cases.

In order to study the distribution and evolution of the transient particle and heat fluxes during edge-localized mode (ELM) bursts, a BOUT++ six-field two-fluid model based on the Braginskii equations with non-ideal physics effects is used to simulate pedestal collapse in divertor geometry. The profiles from the DIII-D H-mode discharge #144382 with fast target heat flux measurements are used as the initial conditions for the simulations. A fluxlimited parallel thermal conduction is used with three values of the flux-limiting coefficient α j , free streaming model with α = 1 j , sheath-limit with α = 0.05 j , and one value in between.The studies show that a 20 times increase in α j leads to ∼6 times increase in the heat flux amplitude to both the inner and outer targets, and the widths of the fluxes are also expanded. The sheath-limit model of flux-limiting coefficient is found to be the most appropriate one, which shows ELM sizes close to the measurements. The evolution of the density profile during the burst of ELMs of DIII-D discharge #144382 is simulated, and the collapse in width and depth of n e are reproduced at different time steps. The growing process of the profiles for the heat flux at divertor targets during the burst of ELMs measured by IRTV (infrared television) is also reproduced by this model. The widths of heat fluxes towards targets are a little narrower, and the peak amplitudes are twice the measurements possibly due to the lack of a model of divertor radiation which can effectively reduce the heat fluxes. The magnetic flutter combined with parallel thermal conduction is found to be able to increase the total heat loss by around 33% since the magnetic flutter terms provide the additional conductive heat transport in the radial direction. The heat flux profile at both the inner and outer targets is obviously broadened by magnetic flutter. The lobe structures near the X-point at LFS are both broadened and elongated due to the magnetic flutter.