Velocity control of Electromagnetic pellet injection based on genetic algorithm in J-TEXT

Electromagnetic pellet injection (EMPI) is a new disruption mitigation system on J-TEXT Tokamak. In the experimental research, there is a great demand for the control accuracy of the injection velocity of the impurity pellet. Therefore, considering the special demand for the armature velocity control of the electromagnetic railgun, a genetic algorithm (GA) based power supply timing optimization method is proposed, which makes the armature move uniformly and linearly through controlling the discharge current waveform smoothly, and makes the pellet reach the preset velocity within the controllable orbit length. In this paper, three strategies of elite retention, mutation and tournament selection crossover are used to generate new population. The results show that not only the current waveform is stable, but also the armature moves in a straight line with almost uniform acceleration. The results show that the optimization method is a practical and effective method to determine the timing of electromagnetic railgun pulse power supply.