Generating of Plasma Discharge Video by Generative Adversarial Network

During the experiment of nuclear fusion, a lot of videos containing plasma discharge are recorded inside Large Helical Device. An observation of the recorded videos of plasma light emission can lead to a new discovery or help to optimize the operation parameter of the experiment. An unusual plasma discharge which may cause a damage on the device, is expected to be foreseen through a prediction method. Due to the shortage of videos having such unusual emission, it is required to generate more videos having similar phenomenon. However, video generation is very challenging as the generated video should have not only similarity in features with the real one but also a plausibility in frame-by-frame transition, especially in the case of plasma discharge videos. Thus, this paper proposes a method to generate a plasma emission video by using Generative Adversarial Network (GAN). It has been confirmed that the proposed generative model can produce a new video having plasma emission phenomenon with a very smooth changing of frames.

[1]  Tatsuya Harada,et al.  Hierarchical Video Generation from Orthogonal Information: Optical Flow and Texture , 2017, AAAI.

[2]  SHOJI Mamoru,et al.  Development of the Real-time lmage Data Acquisition System for Observing the Plasma Dynamic Behavior of LHD Long-pulse Discharges , 2022 .

[3]  Sepp Hochreiter,et al.  GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium , 2017, NIPS.

[4]  Shunta Saito,et al.  Temporal Generative Adversarial Nets with Singular Value Clipping , 2016, 2017 IEEE International Conference on Computer Vision (ICCV).

[5]  Prafulla Dhariwal,et al.  Glow: Generative Flow with Invertible 1x1 Convolutions , 2018, NeurIPS.

[6]  Y. Takeiri.  The Large Helical Device: Entering Deuterium Experiment Phase Toward Steady-State Helical Fusion Reactor Based on Achievements in Hydrogen Experiment Phase , 2018, IEEE Transactions on Plasma Science.

[7]  Antonio Torralba,et al.  Generating Videos with Scene Dynamics , 2016, NIPS.

[8]  M. Shoji.  Radiation Resistant Camera System for Monitoring Deuterium Plasma Discharges in the Large Helical Device , 2020, Plasma and Fusion Research.

[9]  Soumith Chintala,et al.  Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks , 2015, ICLR.

[10]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[11]  Cordelia Schmid,et al.  How good is my GAN? , 2018, ECCV.

[12]  Phillip Isola,et al.  On the "steerability" of generative adversarial networks , 2019, ICLR.

[13]  H. Nakanishi,et al.  Prediction of unusual plasma discharge by using Support Vector Machine , 2021 .