A machine learning approach to identify the universality of solitary perturbations accompanying boundary bursts in magnetized toroidal plasmas.

Experimental observations assisted by 2-D imaging diagnostics on the KSTAR tokamak show that a solitary perturbation (SP) emerges prior to a boundary burst of magnetized toroidal plasmas, which puts forward SP as a potential candidate for the burst trigger. We have constructed a machine learning (ML) model based on a convolutional deep neural network architecture for a statistical study to identify the SP as a boundary burst trigger. The ML model takes sequential signals detected from 19 toroidal Mirnov coils as input and predicts whether each temporal frame corresponds to an SP. We trained the network in a supervised manner on a training set consisting of real signals with manually annotated SP locations and synthetic burst signals. The trained model achieves high performances in various metrics on a test data set. We also demonstrated the reliability of the model by visualizing the discriminative parts of the input signals that the model recognizes. Finally, we applied the trained model to new data from KSTAR experiments, which were never seen during training, and confirmed that the large burst at the plasma boundary that can fatally damage the fusion device always involves the emergence of SP. This result suggests that the SP is a key to understanding and controlling of the boundary burst in magnetized toroidal plasmas.

Solitary perturbations in the steep boundary of magnetized toroidal plasma.

Solitary perturbations (SPs) localized both poloidally and radially are detected within ~100 µs before the partial collapse of the high pressure gradient boundary region (called pedestal) of magnetized toroidal plasma in the KSTAR tokamak device. The SP develops with a low toroidal mode number (typically unity) in the pedestal ingrained with quasi-stable edge-localized mode (QSM) which commonly appears during the inter-collapse period. The SPs have smaller mode pitch and different (often opposite) rotation velocity compared to the QSMs. Similar solitary perturbations are also frequently observed before the onset of complete pedestal collapse, suggesting a strong connection between the SP generation and the pedestal collapse.

Compact ECEI system with in-vessel reflective optics for WEST.

An electron cyclotron emission imaging (ECEI) diagnostic system for WEST (W Environment for Steady state Tokamak) is under development to study the MHD instabilities affected by tungsten impurities. The system will provide 2-D Te fluctuation images (width × height = ∼18 cm × âˆ¼ 34 cm at low field side and ∼13 cm × âˆ¼ 39 cm at high field side) from a poloidal cross section with high spatial (≤1.7 cm) and temporal (≤2 µs) resolutions. While the key concept and electronic structure are similar to that of prior ECEI systems on other tokamak devices such as KSTAR, DIII-D, or ASDEX-U, part of the imaging optics have to be placed inside the vacuum vessel in order to resolve issues on limited installation space and longer beam path to the detector position. The in-vessel optics consisting of two large curvature-radius mirrors are expected to withstand the extreme heating on long-pulse operation scenario (∼1000 s). The out-vessel optical housing is constructed as compact as possible to remove easily from the installation site in case of necessity. Commissioning of the system is scheduled on the second experimental WEST campaign end of 2017.

Ion gyroscale fluctuation measurement with microwave imaging reflectometer on KSTAR.

Ion gyroscale turbulent fluctuations with the poloidal wavenumber kθ ∼ 3 cm-1 have been measured in the core region of the neutral beam (NB) injected low confinement (L-mode) plasmas on Korea superconducting tokamak advanced research. The turbulence poloidal wavenumbers are deduced from the frequencies and poloidal rotation velocities in the laboratory frame, measured by the multichannel microwave imaging reflectometer. Linear and nonlinear gyrokinetic simulations also predict the unstable modes with the normalized wavenumber kθρs ∼ 0.4, consistent with the measurement. Comparison of the measured frequencies with the intrinsic mode frequencies from the linear simulations indicates that the measured ones are primarily due to the E × B flow velocity in the NB-injected fast rotating plasmas.

New compact and efficient local oscillator optic system for the KSTAR electron cyclotron emission imaging system.

Electron cyclotron emission imaging (ECEI) diagnostic on Korean Superconducting Tokamak Advanced Research utilizes quasi-optical heterodyne-detection method to measure 2D (vertical and radial) Te fluctuations from two toroidally separated poloidal cross section of the plasma. A cylindrical lens local oscillator (LO) optics with optical path length (OPL) 2-2.5 m has been used in the current ECEI system to couple the LO source to the 24 vertically aligned array of ECE detectors. For efficient and compact LO optics employing the Powell lens is proposed so that the OPL of the LO source is significantly reduced from ∼2.0 m to 0.4 m with new optics. The coupling efficiency of the LO source is expected to be improved especially at the edge channels. Results from the optical simulation together with the laboratory test of the prototype optics will be discussed in this paper.

Post calibration of the two-dimensional electron cyclotron emission imaging instrument with electron temperature characteristics of the magnetohydrodynamic instabilities.

The electron cyclotron emission imaging (ECEI) instrument is widely used to study the local electron temperature (Te) fluctuations by measuring the ECE intensity IECE ∝ Te in tokamak plasmas. The ECEI measurement is often processed in a normalized fluctuation quantity against the time averaged value due to complication in absolute calibration. In this paper, the ECEI channels are relatively calibrated using the flat Te assumption of the sawtooth crash or the tearing mode island and a proper extrapolation. The 2-D relatively calibrated electron temperature (Te,rel) images are reconstructed and the displacement amplitude of the magnetohydrodynamic modes can be measured for the accurate quantitative growth analysis.

Quasi 3D ECE imaging system for study of MHD instabilities in KSTAR.

A second electron cyclotron emission imaging (ECEI) system has been installed on the KSTAR tokamak, toroidally separated by 1/16th of the torus from the first ECEI system. For the first time, the dynamical evolutions of MHD instabilities from the plasma core to the edge have been visualized in quasi-3D for a wide range of the KSTAR operation (B0 = 1.7∼3.5 T). This flexible diagnostic capability has been realized by substantial improvements in large-aperture quasi-optical microwave components including the development of broad-band polarization rotators for imaging of the fundamental ordinary ECE as well as the usual 2nd harmonic extraordinary ECE.

Toroidal mode number estimation of the edge-localized modes using the KSTAR 3-D electron cyclotron emission imaging system.

A new and more accurate technique is presented for determining the toroidal mode number n of edge-localized modes (ELMs) using two independent electron cyclotron emission imaging (ECEI) systems in the Korea Superconducting Tokamak Advanced Research (KSTAR) device. The technique involves the measurement of the poloidal spacing between adjacent ELM filaments, and of the pitch angle α* of filaments at the plasma outboard midplane. Equilibrium reconstruction verifies that α* is nearly constant and thus well-defined at the midplane edge. Estimates of n obtained using two ECEI systems agree well with n measured by the conventional technique employing an array of Mirnov coils.

Effects of electron-cyclotron-resonance-heating-induced internal kink mode on the toroidal rotation in the KSTAR Tokamak.

It is observed that the magnitude of the toroidal rotation speed is reduced by the central electron cyclotron resonance heating (ECRH) regardless of the direction of the toroidal rotation. The magnetohydrodynamics activities generally appear with the rotation change due to ECRH. It is shown that the internal kink mode is induced by the central ECRH and breaks the toroidal symmetry. When the magnetohydrodynamics activities are present, the toroidal plasma viscosity is not negligible. The observed effects of ECRH on the toroidal plasma rotation are explained by the neoclassical toroidal viscosity in this Letter. It is found that the neoclassical toroidal viscosity torque caused by the internal kink mode damps the toroidal rotation.

Design of the reflective optics for Tore Supra ECEI system.

A 2D electron cyclotron emission (ECE) imaging system for Tore Supra is under design for studying the MHD physics of the magnetically confined plasma such as sawteeth, tearing modes, and turbulent fluctuations. Complex beam path due to the tight access in Tore Supra led to the design of reflective optics made of 6 or more large cylindrical∕flat mirrors. The total path length of the ECE beam is about 11 m, including almost 4 m inside the vacuum vessel. The imaging property of the optics has been estimated using the Gaussian beam simulation and ray transfer analysis. The possible setups for the optical alignment of the diagnostic and the operation scenarios with single- or dual-array measurement system are discussed.

Appearance and dynamics of helical flux tubes under electron cyclotron resonance heating in the core of KSTAR plasmas.

Dual (or sometimes multiple) flux tubes (DFTs) have been observed in the core of sawtoothing KSTAR tokamak plasmas with electron cyclotron resonance heating. The time evolution of the flux tubes visualized by a 2D electron cyclotron emission imaging diagnostic typically consists of four distinctive phases: (1) growth of one flux tube out of multiple small flux tubes during the initial buildup period following a sawtooth crash, resulting in a single dominant flux tube along the m/n=1/1 helical magnetic field lines, (2) sudden rapid growth of another flux tube via a fast heat transfer from the first one, resulting in approximately identical DFTs, (3) coalescence of the two flux tubes into a single m/n=1/1 flux tube resembling the internal kink mode in the normal sawteeth, which is explained by a model of two current-carrying wires confined on a flux surface, and (4) fast localized crash of the merged flux tube similar to the standard sawtooth crash. The dynamics of the DFTs implies that the internal kink mode is not a unique prerequisite to the sawtooth crash, providing a new insight on the control of the sawtooth.

Suppression of edge localized modes in high-confinement KSTAR plasmas by nonaxisymmetric magnetic perturbations.

Edge localized modes (ELMs) in high-confinement mode plasmas were completely suppressed in KSTAR by applying n=1 nonaxisymmetric magnetic perturbations. Initially, the ELMs were intensified with a reduction of frequency, but completely suppressed later. The electron density had an initial 10% decrease followed by a gradual increase as ELMs were suppressed. Interesting phenomena such as a saturated evolution of edge T(e) and broadband changes of magnetic fluctuations were observed, suggesting the change of edge transport by the applied magnetic perturbations.

Effect of synthetic temperature on the dispersion stability of gold nanocolloid produced via electrical explosion of wire.

In this study, gold nanocolloid was produced via the electrical explosion of wire in water, for the purpose of medical treatment. Thus, the use of other additives was avoided to stabilize the gold nanocolloid. The temperature of the water that was to be used for explosion was changed, and its effect on the stability of the gold nanocolloid was investigated. The synthetic temperature was varied from ice temperature to 80 degrees C. The morphology and particle size were studied using a transmission electron microscope. The UV-Vis spectra confirmed the formation of gold nanoparticles in the water. The stability of the gold nanocolloid was estimated using the zeta-potential and Turbiscan methods. The results showed that the synthetic temperature affected the stability of the gold nanocolloid. The TEM images of the gold nanoparticles prepared at low temperatures (0 and 20 degrees C) have several big particles. But, when the synthetic temperature was increased to 80 degrees C, most of the nanoparticles formed a spherical shape, without neck connection. Better stability was obtained in the gold nanocolloid sample prepared at a higher temperature. The gold nanocolloid that was synthesized at 80 degrees C was stable for more than three months, with small sedimentation.

Two-dimensional visualization of growth and burst of the edge-localized filaments in KSTAR H-mode plasmas.

The filamentary nature and dynamics of edge-localized modes (ELMs) in the KSTAR high-confinement mode plasmas have been visualized in 2D via electron cyclotron emission imaging. The ELM filaments rotating with a net poloidal velocity are observed to evolve in three distinctive stages: initial linear growth, interim quasisteady state, and final crash. The crash is initiated by a narrow fingerlike perturbation growing radially from a poloidally elongated filament. The filament bursts through this finger, leading to fast and collective heat convection from the edge region into the scrape-off layer, i.e., ELM crash.

Preparation and stability of gold colloid by electrical explosion of wire in various media.

Gold colloids were prepared by electrical explosion of wire in various media: cold water (0 degrees C), room temperature water (25 degrees C), hot water (80 degrees C), 0.01 M polysorbate surfactant 20 (TW 20) solution, mixture of 0.01 M TW 20 and 0.01 M ascorbic acid. The size distribution of nanoparticles measured by transmission electron microscope was found to shift to a smaller size with a decrease of temperature and a presence of TW 20 surfactant. The multiple light scattering results showed that medium temperature and ambient medium of explosion process is much influence on the stability of colloid. The gold colloid prepared in cold water is unstable in comparison with one prepared in warm and hot water. The best stability of gold colloid obtained with explosion medium of TW 20 and ascorbic acid solution.

A synthetic diagnostic for the evaluation of new microwave imaging reflectometry diagnostics for DIII-D and KSTAR.

The first microwave imaging reflectometry (MIR) system for characterization of fluctuating plasma density has been implemented for the TEXTOR tokamak [H. Park et al., Rev. Sci. Instrum. 75, 3787 (2004)]; an improved MIR system will be installed on DIII-D and KSTAR. The central issue remains in preserving phase information by addressing antenna coupling between the reflection layer and the detector array in the presence of plasma turbulence. A synthetic diagnostic making use of coupled full-wave diffractive codes has been developed in geometries and applied to a variety of optical arrangements. The effectiveness of each scheme is quantitatively compared with respect to the fluctuation levels accessible in the simulation.

Innovations in optical coupling of the KSTAR electron cyclotron emission imaging diagnostic.

The installation of a new electron cyclotron emission imaging diagnostic for the Korea Superconducting Tokamak Advanced Research (KSTAR) is underway, making use of a unique optical port cassette design, which allows placement of refractive elements inside the cryostat region without adverse effects. The result is unprecedented window access for the implementation of a state of the art imaging diagnostic. A dual-array optical design has been developed, capable of simultaneously imaging the high and low field sides of the plasma with independent features of focal plane translation, vertical zoom, and radial channel spacing. The number of translating optics has been minimized by making use of a zoom lens triplet and parabolic plasma facing lens for maximum channel uniformity over a continuous vertical zoom range of 3:1. The simulated performance of this design is presented along with preliminary laboratory characterization data.

Commissioning of electron cyclotron emission imaging instrument on the DIII-D tokamak and first data.

A new electron cyclotron emission imaging diagnostic has been commissioned on the DIII-D tokamak. Dual detector arrays provide simultaneous two-dimensional images of T(e) fluctuations over radially distinct and reconfigurable regions, each with both vertical and radial zoom capability. A total of 320 (20 vertical×16 radial) channels are available. First data from this diagnostic demonstrate the acquisition of coherent electron temperature fluctuations as low as 0.1% with excellent clarity and spatial resolution. Details of the diagnostic features and capabilities are presented.

Development of KSTAR ECE imaging system for measurement of temperature fluctuations and edge density fluctuations.

The ECE imaging (ECEI) diagnostic tested on the TEXTOR tokamak revealed the sawtooth reconnection physics in unprecedented detail, including the first observation of high-field-side crash and collective heat transport [H. K. Park, N. C. Luhmann, Jr., A. J. H. Donné et al., Phys. Rev. Lett. 96, 195003 (2006)]. An improved ECEI system capable of visualizing both high- and low-field sides simultaneously with considerably better spatial coverage has been developed for the KSTAR tokamak in order to capture the full picture of core MHD dynamics. Direct 2D imaging of other MHD phenomena such as tearing modes, edge localized modes, and even Alfvén eigenmodes is expected to be feasible. Use of ECE images of the optically thin edge region to recover 2D electron density changes during L/H mode transitions is also envisioned, providing powerful information about the underlying physics. The influence of density fluctuations on optically thin ECE is discussed.

Data acquisition and processing system of the electron cyclotron emission imaging system of the KSTAR tokamak.

A new innovative electron cyclotron emission imaging (ECEI) diagnostic system for the Korean Superconducting Tokamak Advanced Research (KSTAR) produces a large amount of data. The design of the data acquisition and processing system of the ECEI diagnostic system should consider covering the large data production and flow. The system design is based on the layered structure scalable to the future extension to accommodate increasing data demands. Software architecture that allows a web-based monitoring of the operation status, remote experiment, and data analysis is discussed. The operating software will help machine operators and users validate the acquired data promptly, prepare next discharge, and enhance the experiment performance and data analysis in a distributed environment.