Experimental Study Of Type-Ⅰ ELM Filamentary Structures And Transport On EAST

Following its initial discovery in the ASDEX device in 1982,the high confinement mode(H-mode)has been observed in a range of magnetic confinement fusion facilities.During H-mode discharges,pedestal collapse and recovery caused by periodic crash of edge localized mode(ELM)instability are typically observed.These processes generate a substantial amount of particles and energy from the pedestal region across the magnetic field into the scrape-off layer(SOL),which are eventually transported to the plasma-facing components and divertor target plates,leading to transient rise in heat load.The instantaneous increase in heat load poses a serious threat to the safety and lifetime of the plasma facing components.Therefore,it is quite urgent to conduct research on ELMy physics and develop dedicated diagnostics with high spatiotemporal resolution.This work mainly relies on EAST,a fully superconducting tokamak experimental device,and employs a newly developed high-speed vacuum ultraviolet imaging(VUVI)diagnostic to carry out the experimental research.The primary objective of this study includes two parts: firstly,to optimize the VUVI image data processing method and develop the image data processing program with MATLAB;secondly,to experimentally investigate the characteristics of filamentary structures and transport properties of type-Ⅰ ELMs based on the optimized image processing process on the EAST tokamak.In recent years,a VUVI system has been developed for edge plasma studies on EAST.This diagnostic system is the only critical tool,which can be routinely operated,for studying the two-dimensional filamentary structures of ELMs in the pedestal region of EAST device,and is not influenced by the heating system.The diagnostic measures impurity radiation with a center wavelength of 13.5nm,which mainly emanates from the pedestal region under the current H-mode discharges.The measurement signal of the VUVI exhibits high sensitivity to the fluctuation intensity of electron density and impurity density.By using this diagnostic,the filamentary structures caused by ELMs are successfully observed during H-mode discharges from EAST device.Due to the intricate environment surrounding the device,the signal-to-noise ratio of the raw image data of the VUVI system is relatively low.To improve the efficacy of extracting filamentary structure information from raw data,the present study proposes a noise reduction method that capitalizes on the underlying characteristics of the image data.Specifically,the proposed method combines the Block-matching and 3D filtering(BM3D)algorithm with background subtraction and mathematical morphology processing to effectively process the raw data.In the proposed method,the initial step involves utilizing the BM3 D algorithm to eliminate the noise present in the image,thereby increasing the signal-to-noise ratio.Subsequently,the background difference method is employed to effectively extract the filamentary structures from the processed data.Finally,mathematical morphology is applied to the image to acquire the edge position and pitch angle of the filaments.A comparative analysis of simulation and experimental data demonstrates that the proposed method is highly effective in extracting edge details and accurately determining the pitch angle of filaments as obtained by VUVI diagnostic.Building upon the proposed image processing method,it is used to analyze the filamentary characteristics of type-Ⅰ ELM obtained by the VUVI diagnostic in EAST device and its transport properties in this work.The distribution characteristics of ELMs crash in the toroidal direction are analyzed,and the plasma discharge data of the two shots under different parameters are selected.One shot shows obvious toroidal asymmetry,and its crash duration and intensity are different in different toroidal positions.However,the toroidal distribution caused by the other shot is relatively uniform,and no obvious asymmetry is observed.The toroidal mode number of the filamentary structure derived from the VUVI images increases with the electron density pedestal height.The transport time of ELMs in the SOL under different discharge conditions is analyzed by comparing it with the divertor probe signals.The time delay is estimated by the correlation analysis between these two signals.The result shows that as the signal peak value increases,i.e.,the size of the ELMs increases,the time delay decreases,indicating that the transmission speed of the ELMs in the SOL becomes faster.The result shows the ion convection is dominate in the parallel transport.Additionally,the influence of resonance magnetic perturbation on the ELM size is also analyzed using VUVI data.Under the modulation of the resonance magnetic perturbation,the ELMs amplitude decreases and the frequency increases.When the phase difference of the coil changes periodically,the widths of the filaments also change periodically.