The radial profiles of perpendicular flows in the presence of the magnetic island were firstly measured in the HL-2A tokamak by hopping the work frequency of the Doppler backward scattering reflectometer system along with a two-dimensional electron cyclotron emission imaging diagnostic identifying the island locations. It has been observed that across the O-point cut the perpendicular flow is quite small at the center of the island and strongly enhanced around the boundary of the island, resulting in a large increase of the flow shear in the outer half island, while across the X-point cut the flow is almost flat in the whole island region. Meanwhile it was found that the density fluctuations are generally weakened inside the island. The results indicate that both the perpendicular flow and the density fluctuation level are modulated by the naturally rotating tearing mode near the island boundary. The cross-correlation between the perpendicular flows and the oscillating electron temperature further reveals that the modulation of the perpendicular flow occurs mainly inside and in the vicinity of the island.

Since the last IAEA Fusion Energy Conference in 2018, significant progress of the experimental program of HL-2A has been achieved on developing advanced plasma physics, edge localized mode (ELM) control physics and technology. Optimization of plasma confinement has been performed. In particular, high-N H-mode plasmas exhibiting an internal transport barrier have been obtained (normalized plasma pressure N reached up to 3). Injection of impurity improved the plasma confinement. ELM control using resonance magnetic perturbation (RMP) or impurity injection has been achieved in a wide parameter regime, including Types I and III. In addition, the impurity seeding with supersonic molecular beam injection (SMBI) or laser blow-off (LBO) techniques has been successfully applied to actively control the plasma confinement and instabilities, as well as the plasma disruption with the aid of disruption prediction. Disruption prediction algorithms based on deep learning are developed. A prediction accuracy of 96.8% can be reached by assembling convolutional neural network (CNN). Furthermore, transport resulted from a wide variety of phenomena such as energetic particles and magnetic islands have been investigated. In parallel with the HL-2A experiments, the HL-2M mega-ampere class tokamak was commissioned in 2020 with its first plasma. Key features and capabilities of HL-2M are briefly presented.

Interactions among pedestal shear flows, turbulence, and the formation of the edge transport barrier have been studied in H-mode plasmas of the HL-2A tokamak by multi-channel Doppler reflectometry with high spatiotemporal resolution. Geodesic acoustic mode (GAM) has been observed during the L-I-H transition. It has been observed that the plasma transits into the I-phase when the mean E×B shear flow reaches a critical value. The bi-spectrum analysis has shown that there is a strong interaction between GAM and limit cycle oscillation (LCO), and the energy transfer is from GAM to LCO, suggesting that GAM can assist the L-I transition. The regulation of the edge turbulence by LCOs helps to build the steep pedestal and initialize the confinement improvement of the plasma. It has been found that the mean E×B shear flow is further increased just before the I-H transition, accompanied by the turbulence suppression, leading to the edge transport reduction and the pedestal formation. It has been demonstrated that the increase of the mean E×B shear flow prior to the L-I and I-H transitions is due to the ion diamagnetic component of Er. These results corroborate that the mean E×B shear flow plays a key role in the L-I and I-H transitions.

ITER and to the advanced tokamak operation (e.g. the operation of future HL-2M), such as the access of H-mode, energetic particle physics, edge-localized mode (ELM) mitigation/suppression and disruption mitigation. Since the 2016 Fusion Energy Conference, the HL-2A team has focused on the investigations on the following areas: (i) pedestal dynamics and L-H transition, (ii) techniques of ELM control, (iii) the turbulence and transport, (iv) energetic particle physics. The HL-2A results demonstrated that the increase of mean E × B shear flow plays a key role in triggering L-I and I-H transitions. While the change of E × B flow is mainly induced by the ion pressure gradient. Both mitigation and suppression of ELMs were realized by laser blow-off (LBO) seeded impurity (Al, F e, W). The 30% N e mixture supersonic molecular beam injection (SMBI) seeding also robustly induced ELM mitigation. The ELMs were mitigated by low-hybrid current drive (LHCD). The stabilization of m/n=1/1 ion fishbone activities by electron cyclotron resonance heating (ECRH) was found on the HL-2A. A new m/n=2/1 ion fishbone activity was observed recently, and the modelling indicated that passing fast ions dominantly contribute to the driving of 2/1 fishbone. The non-linear coupling between toroidal Alfven eigenmode (TAE) and tearing mode (TM) leads to the generation of a high frequency mode with the toroidal mode number n=0. The turbulence is modulated by tearing mode when the island width exceeds a threshold and the modulation is localized merely in the inner area of the islands. Meanwhile, turbulence radially spreading takes place across the island region.

Research on beta induced Alfvén eigenmodes (BAEs) driven by energetic ions has been carried out on the HL2A tokamak. The BAEs can be observed in the cases of q min being beyond or below unity. The mode frequencies are around 60-95 kHz. The radial mode structures have been detected by the multichannel microwave reflectometer and soft xray arrays. The results suggest that the BAEs are highly localized at the normalized radius of ρ = 0.07-0.26. Theoretical analyses based on the general fishbonelike dispersion relation and Alfvén mode code support the experimental measurements in both mode frequencies and locations. The BAEs are found to be excited more easily in low electron density discharges. The coexistence of BAEs and reversed shear Alfvén eigenmodes enables the estimation of q min directly from experiments.