Disruption remains to be a serious threat to large tokamaks like the International Thermonuclear Experimental Reactor (ITER). The injection speed of disruption mitigation systems (DMS) driven by high pressure gas is limited by the sound speed of the propellant gas. When extrapolating to ITER-like tokamaks, long overall reaction duration and shallow penetration depth due to low injection speed make it stricter for plasma control system to predict the impending disruptions. Some disruptions with a short warning time may be unavoidable. Thus, a fast time response and high injection speed DMS is essential for large scale devices. The electromagnetic pellet-injection (EMPI) system is a novel massive material injection system aiming to provide rapid and effective disruption mitigation. Based on the railgun concept, EMPI can accelerate the payload to over 1000 m/s and shorten the overall reaction time to a few milliseconds. To verify the injection ability and stability of the EMPI, the prototype injector EMPI-1 has been designed and assembled. The preliminary test has been carried out using a 5.9 g armature to propel a dummy pellet and the results suggest that the EMPI configuration has a great potential to be the DMS of the large scale fusion devices.

As an obstacle in high-performance discharge in future fusion devices, disruptions may do great damages to the reactors through causing strong electromagnetic forces, heat loads and so on. The drift waves in tokamak are illustrated to play essential roles in the confinement performance as well. Depending on the plasma parameters and mode perpendicular wavelength, the mode phase velocity is either in the direction of electron diamagnetic velocity (namely, typical trapped electron mode) or in the direction of ion diamagnetic velocity (namely, the ubiquitous mode). Among them, the ubiquitous mode is directly investigated using gyro-fluid simulation associating with gyro-fluid equations for drift waves in tokamak plasmas. The ubiquitous mode is charactered by the short wavelength and propagates in ion diamagnetic direction. It is suggested that the density gradient is essential for the occurrence of the ubiquitous mode. However, the ubiquitous mode is also influenced by the temperature gradients and other plasma parameters including the magnetic shear and the fraction of trapped electrons. Furthermore, the ubiquitous mode may play essential roles in the turbulent transport. Meanwhile, the relevant parameters are scanned using a great number of electrostatic gyro-fluid simulations. The stability map is taken into consideration with the micro-instabilities contributing to the turbulent transport. The stability valley of the growth rates occurs with the assumption of the normalized temperature gradient equaling to the normalized density gradient.

The generation of runaway electrons (REs) is observed during the low-density helium ohmic plasma discharge in the Experimental Advanced Superconducting Tokamak (EAST). The growth rate of hard x-ray (HXR) is inversely proportional to the line-average density. Besides, the RE generation in helium plasma is higher than that in deuterium plasma at the same density, which is obtained by comparing the growth rate of HXR with the same discharge conditions. The potential reason is the higher electron temperature of helium plasma in the same current and electron density plateau. Furthermore, two Alfvén eigenmodes driven by REs have been observed. The frequency evolution of the mode is not fully satisfied with the Alfvén scaling and when extension of the Alfvén frequency is towards 0, the high frequency branch is ～ 50 kHz. The different spatial position of the two modes and the evolution of the helium concentration could be used to understand deviation between theoretical and experimental observation.

In the experiments of actively triggering plasma disruption by massive gas injection, the externally applied resonant magnetic perturbation has been used to mitigate the hazard of runaway electron (RE). Motivated by the experiment of multimode coupling to suppress REs on J-TEXT, some typical simulation cases with non-ideal MHD with rotation-open discussion (NIMROD) code are carried out to explore the influential mechanism of different relative phases between m/n = 2/1 and m/n = 3/1 magnetic islands on the confinement of REs. Results show that the RE confinement is drastically affected by the relative phase between 2/1 and 3/1 magnetic islands. When the O point phase of 2/1 and 3/1 magnetic islands is toroidal 330°, REs can be effectively lost. The fitting curve of the remaining ratio of REs vs. the relative toroidal phase is predicted to approximate a sine-like function dependence. Further studies indicate that the phase difference between coexisting 2/1 and 3/1 islands can affect the radial transport of impurities. The loss of runaway electrons is closely related to the deposition effect of impurity. The impurity is easier to spread into the core region with smaller poloidal phase difference between the radial velocity of impurity and the impurity quantity of Ar.

Multifaceted asymmetric radiation from the edge (MARFE) movement which can cause density limit disruption is often encountered during high density operation on many tokamaks. Therefore, identifying and predicting MARFE movement is meaningful to mitigate or avoid density limit disruption for the steady-state high-density plasma operation. A machine learning method named random forest (RF) has been used to predict the MARFE movement based on the density ramp-up experiment in the 2022's first campaign of Experimental Advanced Superconducting Tokamak (EAST). The RF model shows that besides Greenwald fraction which is the ratio of plasma density and Greenwald density limit, d$\beta_{\rm p}/$d$ t$, $H_{98}$ and d $W_{\rm mhd}/$d$t$ are relatively important parameters for MARFE-movement prediction. Applying the RF model on test discharges, the test results show that the successful alarm rate for MARFE movement causing density limit disruption reaches $\sim 85%$ with a minimum alarm time of $\sim 40 $ ms and mean alarm time of $\sim 700 $ ms. At the same time, the false alarm rate for non-disruptive and non-density-limit disruptive discharges can be kept below 5%. These results provide a reference to the prediction of MARFE movement in high density plasmas, which can help the avoidance or mitigation of density limit disruption in future fusion reactors.

The effect of tearing modes on the confinement of runaway electrons is studied in Experimental Advanced Superconducting Tokamak (EAST). The general tendency of the radial diffusion coefficient of runaway electrons (REs) D_r is derived based on the time response relation between the tearing modes and runaway electrons. The results indicate that, the magnetic fluctuations of tearing modes will enhance the radial diffusion of runaway electrons when the magnetic island is small. Following the increasing of the magnetic fluctuations of the tearing modes, the formed large magnetic island may weaken the radial diffusion of runaway electrons. The results can be important to understand the confinement of runaway electrons when large magnetic islands exist in the plasma.

Massive gas injection (MGI) is a traditional plasma disruption mitigation method. This method directly injected massive gas into the pre-disruption plasma and had been developed on the Experimental Advanced Superconducting Tokamak (EAST). Different noble gas injection experiments, including He, Ne, and Ar, were performed to compare the mitigation effect of plasma disruption by evaluating the key parameters such as flight time, pre-thermal quench (pre-TQ), and current quench (CQ). The flight time was shorter for low atomic number (Z) gas, and the decrease in flight time by increasing the amount of gas was insignificant. However, both pre-TQ and CQ durations decreased considerably with the increase in gas injection amount. The effect of atomic mass on pre-TQ and CQ durations showed the opposite trend. The observed trend could help in controlling CQ duration in a reasonable area. Moreover, the analysis of radiation distribution with different impurity injections indicated that low Z impurity could reduce the asymmetry of radiation, which is valuable in mitigating plasma disruption. These results provided essential data support for plasma disruption mitigation on EAST and future fusion devices.

Runaway electron current generated during the current quench phase of tokamak disruptions could result in severe damage to future high performance devices. To control and mitigate such runaway electron current, it is important to accurately describe the runaway electron current dominated equilibrium, based on which further stability analysis could be carried out. In this paper, we derive a Grad-Shafranov-like equation solving for the axisymmetric drift surfaces of the runaway electrons instead of the magnetic flux surfaces for the simple case that all runaway electrons share the same parallel momentum. This new equilibrium equation is then numerically solved with simple rectangular wall with ITER-like and MAST-like geometry parameters. The deviation between the drift surfaces and the flux surfaces is readily obtained, and runaway electrons are found to be well confined even in regions with open field lines. The change of the runaway electron parallel momentum is found to result in a horizontal current center displacement without any changes in the total current or the external field. The runaway current density profile is found to affect the susceptibility of such displacement, with flatter profiles result in more displacement by the same momentum change. With up-down asymmetry in the external poloidal field, such displacement is accompanied by a vertical displacement of runaway electron current. It is found that this effect is more pronounced in smaller, compact device and weaker poloidal field cases. The above results demonstrate the dynamics of current center displacement caused by the momentum space change in the runaway electrons, and pave a way for more sophisticated runaway current equilibrium theory in the future with more realistic consideration on the runaway electron momentum distribution. This new equilibrium theory also provides foundation for future stability analysis of the runaway electron current.

The stability (or instability) of finite sized magnetic island could play a significant role in disruption avoidance or disruption mitigation dynamics. Especially, various current and pressure profile modifications, such as the current drive and heating caused by electron cyclotron wave, or the radiative cooling and current expulsion caused by the shattered pellet injection could be applied within the island to modify its stability, thus changing the ensuing dynamics. In this study, we calculate the mode structure modification caused by such profile changes within the island using the perturbed equilibrium approach, thus obtain the change of stability criterion $\varDelta$' and assess the corresponding quasi-linear island stability. The positive helical current perturbation is found to always stabilize the island, while the negative one is found to do the opposite, in agreement with previous results. The pressure bump or hole within the island has a more complicated stability impact. In the small island regime, its contribution is monotonic, with pressure bump that tends to stabilize the island while pressure hole destabilizes it. This effect is relatively weak, though, due to the cancellation of the pressure term's odd parity contribution in the second derivatives of the mode structure. In the large island regime, such cancellation is broken due to the island asymmetry, and the pressure contribution to stability is manifested, which is non-monotonic. The stability analysis in this paper helps to more accurately clarify the expected island response in the presence of profile modifications caused by disruption avoidance or mitigation systems.

Disruption prediction and mitigation is a crucial topic, especially for future large-scale tokamaks, due to disruption's concomitant harmful effects on the devices. On this topic, disruption prediction algorithm takes the responsibility to give accurate trigger signal in advance of disruptions, therefore the disruption mitigation system can effectively alleviate the harmful effects. In the past 5 years, a deep learning-based algorithm is developed in HL-2A tokamak. It reaches a true positive rate of 92.2%, a false positive rate of 2.5% and a total accuracy of 96.1%. Further research is implemented on the basis of this algorithm to solve three key problems, i.e., the algorithm's interpretability, real-time capability and transferability. For the interpretability, HL-2A's algorithm gives saliency maps indicating the correlation between the algorithm's input and output by perturbation analysis. The distribution of correlations shows good coherence with the disruption causes. For the transferability, a preliminary disruption predictor is successfully developed in HL-2M, a newly built tokamak in China. Although only 44 shots are used as the training set of this algorithm, it gives reasonable outputs with the help of data from HL-2A and J-TEXT. For the real-time capacity, the algorithm is accelerated to deal with an input slice within 0.3 ms with the help of some adjustments on it and TFLite framework. It is also implemented into the plasma control system and gets an accuracy of 89.0% during online test. This paper gives a global perspective on these results and discusses the possible pathways to make HL-2A's algorithm a more comprehensive solution for future tokamaks.

Predicting disruptions across different tokamaks is necessary for next generation device. Future large-scale tokamaks can hardly tolerate disruptions at high performance discharge, which makes it difficult for current data-driven methods to obtain an acceptable result. A machine learning method capable of transferring a disruption prediction model trained on one tokamak to another is required to solve the problem. The key is a feature extractor which is able to extract common disruption precursor traces in tokamak diagnostic data, and can be easily transferred to other tokamaks. Based on the concerns above, this paper presents a deep feature extractor, namely, the fusion feature extractor (FFE), which is designed specifically for extracting disruption precursor features from common diagnostics on tokamaks. Furthermore, an FFE-based disruption predictor on J-TEXT is demonstrated. The feature extractor is aimed to extracting disruption-related precursors and is designed according to the precursors of disruption and their representations in common tokamak diagnostics. Strong inductive bias on tokamak diagnostics data is introduced. The paper presents the evolution of the neural network feature extractor and its comparison against general deep neural networks, as well as a physics-based feature extraction with a traditional machine learning method. Results demonstrate that the FFE may reach a similar effect with physics-guided manual feature extraction, and obtain a better result compared with other deep learning methods.