Electron Cyclotron Emission Imaging Research Articles

Implementation of spectral clustering algorithm for automatic identification of plasma coherence patterns

The number of data accumulated by controllable nuclear fusion devices is too large, and a large number of data have not been fully exploited. In such big data processing machine learning can play an important role. Therefore, in this work the spectral clustering method is used to realize the automatic processing of data, which can easily and quickly find the pattern information contained in the data. The discovery of these patterns is of great significance in improving plasma confinement and understanding plasma physics. In addition, in this work the spectral clustering method is applied to the electron cyclotron emission imaging (ECEI), one-dimensional diagnostic system electron cyclotron emissiometer, magnetic probe, soft X-ray, fast radiation (fast bolometer) and other different diagnostic systems on the EAST tokamak device. The sawtooth pattern is identified, the migration of the spectral clustering method is verified, and the problems of poor data processing migration in supervised learning and the need to rely on a large number of labeled data are solved. Finally, in this work, the ECEI and magnetic probe data are used to discover a possible new mode in the time domain and frequency domain respectively, which provides a new idea for exploring new modes.

System-on-chip integrated circuit technology applications on the DIII-D tokamak for multi-field measurements

Several mm-wave diagnostics on the DIII-D tokamak provide multi-scale and multi-dimensional measurements of plasma profile evolution and turbulence fluctuations. Mm-wave fusion plasma diagnostics that adopt system-on-chip integrated circuit technology can provide better space utilization, flexible installation, and improved sensitivity. In order to further extend this technology for additional fusion facilities with a higher toroidal magnetic field, V-band (55–75 GHz) and F-band (90–140 GHz) chips for Microwave Imaging Reflectometer (MIR) and Electron Cyclotron Emission Imaging (ECEI) instruments are developed and tested in the Davis Millimeter Wave Research Center (DMRC). Current measurement data show that correlation between these SoC-based diagnostic instruments with other state-of-the-art diagnostics enables co-located multi-field turbulence fluctuation measurement.

Development of the Q-band ECE imaging system in the large helical device

In this study, we developed the Electron Cyclotron Emission Imaging (ECEI) system with the Q-band in the Large Helical Device (LHD). ECEI measurement makes it possible to obtain the spatiotemporal structure of magnetohydrodynamics (MHD) instabilities in the high-β plasma. Although there were several difficulties for realizing the multi-channelization, such as local oscillator (LO) optics and an expensive high-power LO source, we have solved these problems by developing a Local Integrated Antenna array (LIA) which has an internal LO supply, using a frequency doubler integrated circuit on each channel, instead of a conventional Horn-antenna Mixer Array (HMA) with common LOs. In addition, we have made some improvements to enhance the quality of the measurement signal. First, we developed and introduced notch filters that prevent the strong Electron Cyclotron Resonance Heating (ECRH) stray signal from being mixed into the measurement circuit. Second, the position of the doubler built in the printed circuit board was reconsidered to prevent the mixing of higher harmonic components into the mixer. Also, we have adopted the Logarithmic detector (LOG detector) to deal with the wide dynamic range of the plasma fluctuation. After these improvements, for the first time, we could successfully obtain the initial results of the two-dimensional temperature distribution and its fluctuation distribution in the LHD.

High level of integration of front-end imaging optics system for electron cyclotron emission imaging diagnostics on the DIII-D tokamak

The optics system of the Electron Cyclotron Emission Imaging (ECEI) diagnostics on the DIII-D tokamak has been developed and installed for high-resolution temperature fluctuations measurement in the previous decade. The System-on-Chip approach upgrade was finished in June 2019 with high integrated receiver array with 20 individual modules. In this paper, we design new optics with high-level integration. In the new optics system, there will be two mirrors from plasma to receiver array, including one replaceable curved mirror out of the vacuum and one fixed curved mirror in the vessel. There are two options with 1:1 and 1:2 magnifications for high poloidal resolution and larger poloidal window modes. The new optics system has multiple improvements, including lighter weight, smaller size, more flexible installation and alignment. The synthetic ECEI module is used to evaluate the new optics system.

Observation of off-axis sawtooth oscillations during the presence of nonlinear mode couplings in HL-2A NBI heated plasmas

Off-axis sawtooth oscillations during the presence of nonlinear coupling between toroidal Alfvén eigenmodes (TAEs) and tearing mode are firstly observed in the plasma without two q = 2 rational surfaces on HL-2A tokamak. Those low frequency oscillations always take place followed by nonlinear interaction processes and the off-axis features have been confirmed by multiple diagnostics, such as soft x-ray arrays and electron cyclotron emission imaging systems. A possible physical mechanism is proposed for interpretation of those off-axis oscillations. That is, nonlinear interaction between TAEs and tearing mode causes significant losses of fast ions though the spacial overlapped global mode structure, then makes a drop in torque and leads to rotation braking or even mode locking. When magnetic fluctuation becomes large enough during braking or locking, the resulting high amplitude saturated mode leads to subsequent or immediate collapse at the q = 2 rational surface.

Machine learning application to predict the electron temperature on the J-TEXT tokamak

The reliability of diagnostic systems in tokamak plasma is of great significance for physics researches or fusion reactor. When some diagnostics fail to detect information about the plasma status, such as electron temperature, they can also be obtained by another method: fitted by other diagnostic signals through machine learning. The paper herein is based on a machine learning method to predict electron temperature, in case the diagnostic systems fail to detect plasma temperature. The fully-connected neural network, utilizing back propagation with two hidden layers, is utilized to estimate plasma electron temperature approximately on the J-TEXT. The input parameters consist of soft x-ray emission intensity, electron density, plasma current, loop voltage, and toroidal magnetic field, while the targets are signals of electron temperature from electron cyclotron emission and x-ray imaging crystal spectrometer. Therefore, the temperature profile is reconstructed by other diagnostic signals, and the average errors are within 5%. In addition, generalized regression neural network can also achieve this function to estimate the temperature profile with similar accuracy. Predicting electron temperature by neural network reveals that machine learning can be used as backup means for plasma information so as to enhance the reliability of diagnostics.

System-on-chip upgrade of millimeter-wave imaging diagnostics for fusion plasma.

Monolithic, millimeter wave "system-on-chip" technology has been employed in chip heterodyne radiometers in a newly developed Electron Cyclotron Emission Imaging (ECEI) system on the DIII-D tokamak for 2D electron temperature and fluctuation diagnostics. The system employs 20 horn-waveguide receiver modules each with customized W-band (75-110GHz) monolithic microwave integrated circuit chips comprising a W-band low noise amplifier, a balanced mixer, a ×2 local oscillator (LO) frequency doubler, and two intermediate frequency amplifier stages in each module. Compared to previous quasi-optical ECEI arrays with Schottky mixer diodes mounted on planar antennas, the upgraded W-band array exhibits >30 dB additional gain and 20× improvement in noise temperature; an internal eight times multiplier chain is used to provide LO coupling, thereby eliminating the need for quasi-optical coupling. The horn-waveguide shielding housing avoids out-of-band noise interference on each module. The upgraded ECEI system plays an important role for absolute electron temperature and fluctuation measurements for edge and core region transport physics studies. An F-band receiver chip (up to 140GHz) is under development for additional fusion facilities with a higher toroidal magnetic field. Visualization diagnostics provide multi-scale and multi-dimensional data in plasma profile evolution. A significant aspect of imaging measurement is focusing on artificial intelligence for science applications.

Design of microwave broadband CMOS transmitter and receiver circuits for MIR and ECEI plasma diagnostics.

To efficiently determine the plasma electron density fluctuations using the MIR diagnostic technique, a 55-75GHz 65 nm-CMOS transmitter has been developed where four separate intermediate frequency (IF) signals are up-converted, amplified, and then combined to generate an 8-tone RF output; a broadband 90 nm-CMOS receiver has also been constructed, which consists of an RF-low noise amplifier (LNA), mixer, and IF amplifier. The circuits and their corresponding modules will soon be deployed on the DIII-D and NSTX-U fusion devices. A 110-140GHz 65 nm-CMOS receiver has also been designed, which is suitable for measuring the deep-core temperature fluctuations in the DIII-D tokamak using the electron cyclotron emission imaging diagnostic system. In addition to the RF-LNA/balun, mixer, and IF amplifier, an LO balun/tripler and driving amplifier are now included in this highly integrated circuit chip. By adopting the microwave and millimeter-wave system-on-chip concept in the front-end system design, this paper demonstrates that compact transmitter and receiver modules can be easily built, which, in turn, facilitates array implementation and maintenance.

Noise suppression for MHD characterization with electron cyclotron emission imaging 1D technique

Significant noise suppression for magnetohydrodynamics (MHD) mode characterization in the spatial and spectral domain is achieved by processing two-dimensional (2D) electron cyclotron emission imaging (ECEI) data with a one-dimensional (1D) ECEI technique using a short time window (). The technique is applied to detect toroidal Alfven eigenmodes (TAEs) in the temporal spectrum and fit their radial envelope using the data from the DIII-D tokamak W-band 2D ECEI system. Using the data length (time window) of only 1 , the 1D ECEI can clearly detect the TAEs (∼100 ) on the spectrum, while similar spectrum quality requires ∼10 data length with the cross power spectrum between two midplane ECEI channels. The 1D ECEI technique also effectively avoids biased fitting when resolving the fine structure of the TAE’s radial envelope. The radially spatial resolution of 1D ECEI is constrained by the finite ECE radiation volume of the ECEI receiver. With forward radiation modeling, we find the DIII-D ECEI system can sensitively measure the even parity MHD activities, for which the mode width is >15 mm, and tearing modes (odd parity MHD activities), for which the island full width is >30 mm.

Experimental investigation of a partial collapse during the sawtooth instability in EAST

ABSTRACT In this article, partial and full collapses of sawtooth activity are investigated with the high-resolution two-dimensional electron cyclotron emission (ECE) Imaging diagnostic in EAST. The partial collapse in normal sawtooth behavior is detected during the precursor phase of an L-mode plasma heated by ICRF. The electron heat diffusivity is determined with the time-to-peak analysis. The experimental results revealed that heat diffusivity suppresses at the time of the partial collapse. As a result, the partial collapse process interrupts and the small amount of heat releases outward. The hot core survives and is enclosed by the cold magnetic island. The inversion radii are estimated with the help of sawtooth and the only single inversion radius is found during the entire collapse process.

Q-band high-performance notch filters at 56 and 77GHz notches for versatile fusion plasma diagnostics.

A six-pole Q-band waveguide filter with a notch frequency above the Q-band has been developed for plasma diagnostics. The previous paper [Nishiura et al., J. Instrum. 10, C12014 (2015)] reported that the notch frequency exists within the standard band. In this study, the newly required notch filter extends the function, which prevents a thorny wave from being mixed into an instrument beyond the standard bandwidth of the waveguide. The mode control technique for cavities realizes a deep and sharp filter shape for Q-band notch filters with 56 and 77GHz notches, respectively. The former filter has an attenuation more than 50 dB at 56.05GHz and a bandwidth of 1.1GHz at -3 dB. The latter filter has an attenuation more than 55 dB at 76.95GHz and a bandwidth of 1.6GHz at -3 dB. The electron cyclotron emission imaging and the electron cyclotron emission (ECE) diagnostics for the Q-band implemented a pair of the fabricated filters and demonstrated the ECE measurement successfully in the intense stray radiation from a 56GHz gyrotron.

Integrated package of electron cyclotron emission imaging data processing and forward modeling in OMFIT.

An Electron Cyclotron Emission Imaging (ECEI) data analysis module has been developed for the OMFIT platform to accommodate the needs of users at the DIII-D tokamak for physics applications. The user can easily access the ECEI spatial observation windows in the plasma that are calculated based on the automatically retrieved hardware setup and available DIII-D equilibria, perform spectral analysis, and obtain 2D electron temperature fluctuation images. The module provides a powerful data post-processing package for extracting important physics parameters from the 2D measurements, including the radial structure and poloidal mode number of Alfven eigenmodes, as well as the frequency-vs-wavenumber dispersion relationship of broadband MHD. The module propagates characterized synthetic fluctuations for the user, so one can perform forward modeling tasks with simple analytical fluctuations.

A Framework for International Collaboration on ITER Using Large-Scale Data Transfer to Enable Near-Real-Time Analysis

The global nature of the ITER project along with its projected approximately petabyte-per-day data generation presents not only a unique challenge but also an opportunity for the fusion community to rethink, optimize, and enhance our scientific discovery process. Recognizing this, collaborative research with computational scientists was undertaken over the past several years to create a framework for large-scale data movement across wide-area networks to enable global near-real-time analysis of fusion data. This would broaden the available computational resources for analysis/simulation and increase the number of researchers actively participating in experiments. An official demonstration of this framework for fast, large data transfer and real-time analysis was carried out between the KSTAR tokamak in Daejeon, Korea, and Princeton Plasma Physics Laboratory (PPPL) in Princeton, New Jersey. Streaming large data transfer, with near-real-time movie creation and analysis of the KSTAR electron cyclotron emission imaging data, was performed using the Adaptable Input Output (I/O) System (ADIOS) framework, and comparisons were made at PPPL with simulation results from the XGC1 code. These demonstrations were made possible utilizing an optimized network configuration at PPPL, which achieved over 8.8 Gbps (88% utilization) in throughput tests from the National Fusion Research Institute to PPPL. This demonstration showed the feasibility for large-scale data analysis of KSTAR data and provides a nascent framework to enable use of globally distributed computational and personnel resources in pursuit of scientific knowledge from the ITER experiment.

Role of fast-ion transport manipulating safety factor profile in KSTAR early diverting discharges

Magnetohydrodynamics (MHD) activity is analyzed for a particular KSTAR discharge category, the main features of which are a high- and a broad q profile. An interesting finding in this scenario is that a steady-state high- and broad q profile has been constructed even without a strong off-axis current drive scheme. Time traces of magnetics spectrograms and electron cyclotron emission imaging (ECEI) clearly show that Alfvénic type MHD activity appears as a high- and broad q profile is formed. Conversely, this activity disappears as the current profiles evolve toward a low- profile. In this study, the effect of MHD modes and accompanying fast-particle transport on the formation of various current profiles is explored. The frequency patterns of the magnetics spectrograms and ECEI coherence are compared using a NOVA analysis to determine which series of Alfvénic modes are active. Then, the kick-model is applied to potential modes, and used to estimate how much fast-ions are transported and beam driven current profiles are varied. By comparing the reconstructed total current profile from the kinetic EFIT with the beam-driven current profile estimated from the kick model, it is determined how energetic particle transport is responsible for maintaining the safety factor profile and which q profile can provide high βN operation.

W-band system-on-chip electron cyclotron emission imaging system on DIII-D.

Monolithic, millimeter-wave "system-on-chip" (SoC) technology has been employed in heterodyne receiver integrated circuit radiometers in a newly developed Electron Cyclotron Emission Imaging (ECEI) system on the DIII-D tokamak for 2D electron temperature profile and fluctuation evolution diagnostics. A prototype module operating in the E-band (72 GHz-80 GHz) was first employed in a 2 × 10 element array that demonstrated significant improvements over the previous quasi-optical Schottky diode mixer arrays during the 2018 operational campaign of the DIII-D tokamak. For compatibility with International Thermonuclear Experimental Reactor relevant scenarios on DIII-D, the SoC ECEI system was upgraded with 20 horn-waveguide receiver modules. Each individual module contains a University of California Davis designed W-band (75 GHz-110 GHz) receiver die that integrates a broadband low noise amplifier, a double balanced down-converting mixer, and a ×4 multiplier on the local oscillator (LO) chain. A ×2 multiplier and two IF amplifiers are packaged and selected to further boost the signal strength and downconvert the signal frequency. The upgraded W-band array exhibits >30 dB additional gain and 20× improvement in noise temperature compared with the previous Schottky diode radio frequency mixer input systems; an internal 8 times multiplier chain is used to bring down the LO frequency below 12 GHz, thereby obviating the need for a large aperture for quasi-optical LO coupling and replacing it with coaxial connectors. Horn-waveguide shielding housing avoids out-of-band noise interference on each individual module. The upgraded ECEI system plays an important role for absolute electron temperature evolution and fluctuation measurements for edge and core region transport physics studies.

A spectral clustering based method for automatic plasma pattern identification in electron cyclotron emission imaging on EAST Tokamak

Electron cyclotron emission imaging (ECEI) diagnostic system of Experimental Advanced Superconducting Tokamak (EAST) can provide two-dimensional electron temperature fluctuation on the cross section of the EAST Tokamak. A major task in ECEI data analysis is to recognize patterns in ECEI signals. In this paper, we present a machine learning method to automatically identify the temporal and spacial location of plasma patterns inside ECEI data of EAST tokamak. The method is based on spectral clustering of ECEI signals of different channels. It is applied to ECEI data of 5 discharges obtained from EAST tokamak and the results can verify the effectiveness of our method. Theoretically it can be applied to recognize patterns in many temporal and spacial diagnostics.

A Constant Corrugation Circular Waveguide for High-Pass Signal Diagnostics in ECEI System at 75–110 GHz

In this work, a high-pass waveguide filter with constant circular corrugations is presented. The proposed design allows propagation of HE11 mode which has similar Gaussian characteristics of the signals present in a fusion reactor. The device is designed to operate in W-band, i.e., 75–110 GHz, and it generates a wide passband of approximately 30 GHz. The internal corrugations in the proposed design are kept constant which makes fabrication relatively simpler. The proposed high-pass metallic waveguide is numerically analyzed, simulated, and fabricated for experimental validation. The measured transmission co-efficient (S21) shows good agreement with the simulated result and passband of 28 GHz was obtained. The high-pass filter can be effectively used in the electron cyclotron emission imaging (ECEI) system used in fusion plasma experiments for signal diagnostics.

Validation of the BEAMS3D neutral beam deposition model on Wendelstein 7-X

The neutral beam deposition model in the BEAMS3D code is validated against neutral beam attenuation data from Wendelstein 7-X (W7-X). A set of experimental discharges where the neutral beam injection system of W7-X was utilized were reconstructed. These discharges scanned the magnetic configurations and plasma densities of W7-X. The equilibrium reconstructions were performed using STELLOPT which calculates three-dimensional self-consistent ideal magnetohydrodynamic equilibria and kinetic profiles. These reconstructions leveraged new capabilities to incorporate electron cyclotron emission and x-ray imaging diagnostics in the STELLOPT code. The reconstructed equilibria and profiles served as inputs for BEAMS3D calculations of neutral beam deposition in W7-X. It is found that if reconstructed kinetic profiles are utilized, good agreement between measured and simulated beam attenuation is found. As deposition models provide initial conditions for fast-ion slowing down calculations, this work provides a first step towards validating our ability to predict fast ion confinement in stellarators.

Deep convolutional neural networks for multi-scale time-series classification and application to tokamak disruption prediction using raw, high temporal resolution diagnostic data

In this paper, we discuss recent advances in deep convolutional neural networks (CNNs) for sequence learning, which allow identifying long-range, multi-scale phenomena in long sequences, such as those found in fusion plasmas. We point out several benefits of these deep CNN architectures, such as not requiring experts such as physicists to hand-craft input data features, the ability to capture longer range dependencies compared to the more common sequence neural networks (recurrent neural networks like long short-term memory networks), and the comparative computational efficiency. We apply this neural network architecture to the popular problem of disruption prediction in fusion energy tokamaks, utilizing raw data from a single diagnostic, the Electron Cyclotron Emission imaging (ECEi) diagnostic from the DIII-D tokamak. Initial results trained on a large ECEi dataset show promise, achieving an F1-score of ∼91% on individual time-slices using only the ECEi data. This indicates that the ECEi diagnostic by itself can be sensitive to a number of pre-disruption markers useful for predicting disruptions on timescales for not only mitigation but also avoidance. Future opportunities for utilizing these deep CNN architectures with fusion data are outlined, including the impact of recent upgrades to the ECEi diagnostic.

Quasi-optical electron cyclotron emission imaging diagnostic advancements on the J-TEXT tokamak

Abstract The new quasi-optical electron cyclotron emission imaging (ECEI) diagnostic system on the J-TEXT tokamak has been developed and characterized. The quasi-optical system consists of two subsystems: the major imaging optics system for plasma imaging and the local oscillator (LO) optics system to drive the mixers in the antenna array. For the updated demands, some modifications and adaptations are accomplished on the major imaging optics. The vertical zoom ratio, focus capability, and field curvature have been measured on the J-TEXT facility, with good matching with the simulation results. For the various experimental requirements on J-TEXT, the remote control system is adopted for the major imaging optics. As for LO optics systems, spherical LO optics and hyperbolic LO optics have demonstrated high performance. Uniform power deposition has been optimized to drive all mixers and minimize insertion loss.