Recent progress on deep learning-based disruption prediction algorithm in HL-2A tokamak

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