Measuring the electron temperature and identifying plasma detachment using machine learning and spectroscopy.

A machine learning approach has been implemented to measure the electron temperature directly from the emission spectra of a tokamak plasma. This approach utilized a neural network (NN) trained on a dataset of 1865 time slices from operation of the DIII-D tokamak using extreme ultraviolet/vacuum ultraviolet emission spectroscopy matched with high-accuracy divertor Thomson scattering measurements of the electron temperature, Te. This NN is shown to be particularly good at predicting Te at low temperatures (Te < 10 eV) where the NN demonstrated a mean average error of less than 1 eV. Trained to detect plasma detachment in the tokamak divertor, a NN classifier was able to correctly identify detached states (Te < 5 eV) with a 99% accuracy (an F1 score of 0.96) at an acquisition rate 10× faster than the Thomson scattering measurement. The performance of the model is understood by examining a set of 4800 theoretical spectra generated using collisional radiative modeling that was also used to predict the performance of a low-cost spectrometer viewing nitrogen emission in the visible wavelengths. These results provide a proof-of-principle that low-cost spectrometers leveraged with machine learning can be used to boost the performance of more expensive diagnostics on fusion devices and be used independently as a fast and accurate Te measurement and detachment classifier.

Laser calibration of the DIII-D coherence imaging system.

In this paper, we describe an in situ calibration technique for Coherence Imaging Spectroscopy (CIS) that measures 2-D images of ion flows on DIII-D. A low power CW diode laser that is tuneable in the range 464-468 nm along with a precision wavemeter (10-5 nm resolution) is used to characterize the interferometer phase as a function of wavelength in the region of C iii (465 nm) and He ii (468 nm). The interferometer is stabilized both mechanically and thermally to minimize drift during the calibration. Optical stirring and an integration sphere are used to obtain spatially uniform calibration images. The quality of the calibration data enables a measurement of phase versus wavelength over approximately 10 fringes of the interferometer. These coefficients can also be related to the geometry of the optics and the birefringent crystal of the interferometer. On DIII-D, the integration sphere with the laser light is inserted into the CIS optical system between shots and the laser image and wavelength are automatically recorded, providing a zero velocity reference.

Tomographic analysis of tangential viewing cameras (invited).

Many tokamaks now use visible light cameras to observe plasma-wall interactions and integrated line emission. The DIII-D coherence imaging spectroscopy diagnostic cameras image interferograms that encode line integrated velocity. By modeling the 2D camera image pixels as line of sight integrals through an axisymmetric discrete grid, it is possible to do tomographic analysis to determine the local plasma line emissivity and parallel velocity. Methods to solve the inverse problem posed by these tangential viewing cameras are presented. The inversion begins with calculation of the sparse response matrix that encompasses all the geometry and diagnostic information and reduces the process of image formation to a sparse matrix-vector multiply. This work includes techniques for determining the detailed geometry of the camera views and methods for handling physical quantities that vary spatially. Additionally, the size of the response matrix has driven the development of capability to distribute the coarse parallel calculation across a heterogeneous cluster of computers on the Energy Sciences Network. Iterative techniques are then used to solve the sparse matrix-vector linear system.

Verification of Doppler coherence imaging for 2D ion velocity measurements on DIII-D.

Coherence Imaging Spectroscopy (CIS) has emerged as a powerful tool for investigating complex ion phenomena in the boundary of magnetically confined plasma devices. The combination of Fourier-transform interferometry and high-resolution fast-framing cameras has made it possible to make sensitive velocity measurements that are also spatially resolved. However, this sensitivity makes the diagnostic vulnerable to environmental effects including thermal drifts, vibration, and magnetic fields that can influence the velocity measurement. Additionally, the ability to provide an absolute calibration for these geometries can be impacted by differences in the light-collection geometry between the plasma and reference light source, spectral impurities, and the presence of thin-films on in-vessel optics. This paper discusses the mitigation of these effects and demonstration that environmental effects result in less than 0.5 km/s error on the DIII-D CIS systems. A diagnostic comparison is used to demonstrate agreement between CIS and traditional spectroscopy once tomographic artifacts are accounted for.

E×B Flux Driven Detachment Bifurcation in the DIII-D Tokamak.

A bifurcative step transition from low-density, high-temperature, attached divertor conditions to high-density, low-temperature, detached divertor conditions is experimentally observed in DIII-D tokamak plasmas as density is increased. The step transition is only observed in the high confinement mode and only when the B×∇B drift is directed towards the divertor. This work reports for the first time a theoretical explanation and numerical simulations that qualitatively reproduce this bifurcation and its dependence on the toroidal field direction. According to the model, the bifurcation is primarily driven by the interdependence of the E×B-drift fluxes, divertor electric potential structure, and divertor conditions. In the attached conditions, strong potential gradients in the low field side (LFS) divertor drive E×B-drift flux towards the high field side divertor, reinforcing low density, high temperature conditions in the LFS divertor leg. At the onset of detachment, reduction in the potential gradients in the LFS divertor leg reduce the E×B-drift flux as well, such that the divertor plasma evolves nonlinearly to high density, strongly detached conditions. Experimental estimates of the E×B-drift fluxes, based on divertor Thomson scattering measurements, and their dependence on the divertor conditions are qualitatively consistent with the numerical predictions. The implications for divertor power exhaust and detachment control in the next step fusion devices are discussed.