Multi-scale interactions have been observed recently in the HL-2A core NBI plasmas, including the synchronous coupling between kink mode and tearing mode, nonlinear couplings of TAE/BAE and TM near surface, AITG/KBM/BAE and kink mode near surface, and between kink mode and high-frequency turbulence. Experimental results suggest that several couplings can exist simultaneously, Alfvenic fluctuations have an important contribution to the high-frequency turbulence spectra, and the couplings reveal the electromagnetic character. Multi-scale interactions via the nonlinear modulation process maybe enhance plasma transport and trigger sawtooth-crash onset.

The oscillations of poloidal plasma flows induced by radially sheared zonal flows are investigated by newly developed correlation Doppler reflectometers in the HL-2A tokamak. The non-disturbing diagnostic allows one to routinely measure the rotation velocity of turbulence, and hence the radial electric field fluctuations. With correlation Doppler reflectometers, a three-dimensional spatial structure of geodesic acoustic mode (GAM) is surveyed, including the symmetric feature of poloidal and toroidal E r fluctuations, the dependence of GAM frequency on radial temperature and the radial propagation of GAMs. The co-existence of low-frequency zonal flow and GAM is presented. The temporal behaviors of GAM during ramp-up experiments of plasma current and electron density are studied, which reveal the underlying damping mechanisms for the GAM oscillation level.

Increasing the plasma density is one of the key methods in achieving an efficient fusion reaction. High-density operation is one of the hot topics in tokamak plasmas. Density limit disruptions remain an important issue for safe operation. An effective density limit disruption prediction and avoidance system is the key to avoid density limit disruptions for long pulse steady state operations. An artificial neural network has been developed for the prediction of density limit disruptions on the J-TEXT tokamak. The neural network has been improved from a simple multi-layer design to a hybrid two-stage structure. The first stage is a custom network which uses time series diagnostics as inputs to predict plasma density, and the second stage is a three-layer feedforward neural network to predict the probability of density limit disruptions. It is found that hybrid neural network structure, combined with radiation profile information as an input can significantly improve the prediction performance, especially the average warning time (T warn ). In particular, the T warn is eight times better than that in previous work (Wang et al 2016 Plasma Phys. Control. Fusion 58 055014) (from 5 ms to 40 ms). The success rate for density limit disruptive shots is above 90%, while, the false alarm rate for other shots is below 10%. Based on the density limit disruption prediction system and the real-time density feedback control system, the on-line density limit disruption avoidance system has been implemented on the J-TEXT tokamak.

The impact of impurity ions on a pedestal has been investigated in the HL-2A Tokamak, at the Southwestern Institute of Physics, Chengdu, China. Experimental results have clearly shown that during the H-mode phase, an electromagnetic turbulence was excited in the edge plasma region, where the impurity ions exhibited a peaked profile. It has been found that double impurity critical gradients are responsible for triggering the turbulence. Strong stiffness of the impurity profile has been observed during cyclic transitions between the I-phase and H-mode regime. The results suggest that the underlying physics of the self-regulated edge impurity profile offers the possibility for an active control of the pedestal dynamics via pedestal turbulence.

In HL-2A and J-TEXT ohmic confinement regimes, an electrostatic turbulence with quasi-coherent characteristics in spectra of density fluctuations was observed by multi-channel microwave reflectometers. These quasi-coherent modes (QCMs) were detectable in a large plasma region (r/a∼0.3−0.8). The characteristic frequencies of QCMs were in the range of 30–140 kHz. The mode is rotated in the electron diamagnetic direction. In the plasmas with QCMs, trapped electron mode (TEM) was predicted to be unstable by gyrokinetic simulations. The combined experimental results show that the TEM is survived in the linear ohmic confinement regime of plasmas. The quasi-coherent TEM was replaced by broad-band fluctuations when the plasma transits from linear to saturated ohmic confinement regime. The observation was strongly related to the turbulence transition from TEM to ion temperature gradient mode. A critical gradient threshold for TEM excitation in electron temperature gradient was directly found. The effect of TEM on density profile peaking was presented.

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.

The resonant interaction between energeticparticles and tearing mode is an unresolved physics issue at present. It is found for the first time in tokamaks that an unstable tearing mode with slowly rotating m/n = 2/1 helicity, where m/n represent poloidal/toroidal mode numbers, interacts with energeticions and results in amplitudebursting/frequencychirping fishbonelike activities. Nonlinear hybrid kineticMHD simulations with M3DK code prove that the copassing energeticions are responsible for the drive of tearing modes, and the wave particle resonance condition is satisfied at ω φ − 2ω θ − ω = 0, where ω φ , ω θ and ω are the toroidal, poloidal angular frequencies of energeticions and the mode frequency respectively. These findings can help the understanding of tearing mode induced energeticparticle loss and particle acceleration during the tearing mode reconnection in laboratory and space plasmas.

In HL-2A, the characteristics of the edge plasma instabilities and their effects on the dynamical evolution of the pedestal in H-mode plasmas have been investigated. In the edge pedestal region with steep pressure gradient, a quasi-coherent mode (QCM) has been observed in density fluctuations with a frequency range of 50-100 kHz. It appears during the edge localized mode (ELM)-free period after the L-H transition and prior to the first ELM. A threshold in the pedestal density gradient has been identified for the excitation of this mode. The QCM can also be observed during inter-ELM periods. It is excited early in the inter-ELM period, and disappears when the ELM onset starts. The radial wave-number of the mode is estimated with two radially separated reflectometers. It shows that the mode is radially propagating inward. The poloidal wave number estimated with the Langmuir probes is k θ ~ 0.43 cm −1 . The mode propagates poloidally in the electron diamagnetic direction in the plasma frame. The toroidal mode number, deduced from Mirnov signals, is n ~ 7. The corresponding poloidal mode number is m ~ 21 according to the local safety factor value. The analysis for the dynamical evolution of the pedestal during the ELM cycle clearly shows that the mode is excited before the ELM onset. During and after the ELM crash, the mode disappears. It suggests that the QCM is driven by the pedestal density gradient, and the mode in return regulates the pedestal density evolution.