TCV Tokamak Research Articles

Improved Plasma Vertical Position Control on TCV Using Model-Based Optimized Controller Synthesis

Elongated plasmas lead to improved performance in tokamaks but make the plasma prone to vertical instability, which requires active feedback control, a critical issue for future fusion reactors. Vertical control was optimized for the TCV tokamak by applying modern control theory to electromagnetic models for the plasma-vessel-coils dynamics. Two different optimal combinations of poloidal field coils for vertical control actuation are derived from linear plasma response models and used on different timescales for controlling the plasma vertical position. On fast timescales, the priority is input minimization, while on long timescales position control is designed to be compatible with shape control. A structured H-infinity design extending classical H-infinity to fixed-structure control systems was subsequently applied to obtain an optimized controller using all available coils for position control. Closed-loop performance improvement was demonstrated in dedicated TCV experiments, showing a reduction of input requirement for stabilizing the same plasma, thus reducing the risk of power supply saturation and consequent loss of vertical control. This novel algorithm is adaptable to different plasma equilibria as it is designed for model-based automated coil selection and controller tuning, thus avoiding extensive experimental gain scans when performing plasma discharges in TCV. The presented technique is general and can be applied to any present tokamak with independent coils or for the design of future tokamak magnetic control systems.

The GBS code for the self-consistent simulation of plasma turbulence and kinetic neutral dynamics in the tokamak boundary

A new version of GBS (Ricci et al. (2012) [27]; Halpern et al. J. Comput. Phys. 315 (2016) 388-408; Paruta et al. (2018) [11]) is described. GBS is a three-dimensional, flux-driven, global, two-fluid turbulence code developed for the self-consistent simulation of plasma turbulence and kinetic neutral dynamics in the tokamak boundary. In the new version presented here, the simulation domain is extended to encompass the whole plasma volume, avoiding an artificial boundary with the core, hence retaining the core-edge-SOL interplay. A toroidal coordinate system is introduced to increase the code flexibility, allowing for the simulation of arbitrary magnetic configurations (e.g. single-null, double-null and snowflake configurations), which can also be the result of the equilibrium reconstruction of an experimental discharge. The implementation of a new iterative solver for the Poisson and Ampère equations is presented, leading to a remarkable speed-up of the code with respect to the use of direct solvers, therefore allowing for efficient electromagnetic simulations that avoid the use of the Boussinesq approximation. The self-consistent kinetic neutral model, initially developed for limited configurations, is ported to the magnetic configurations considered by the present version of GBS and carefully optimized. A new MPI parallelisation is implemented to evolve the plasma and neutral models in parallel, thus improving the code scalability. The numerical implementation of the plasma and neutral models is verified by means of the method of manufactured solutions. As an example of the simulation capabilities of the new version of GBS, a simulation of a TCV tokamak discharge is presented.

First-Principles Density Limit Scaling in Tokamaks Based on Edge Turbulent Transport and Implications for ITER.

A first-principles scaling law, based on turbulent transport considerations, and a multimachine database of density limit discharges from the ASDEX Upgrade, JET, and TCV tokamaks, show that the increase of the boundary turbulent transport with the plasma collisionality sets the maximum density achievable in tokamaks. This scaling law shows a strong dependence on the heating power, therefore predicting for ITER a significantly larger safety margin than the Greenwald empirical scaling [Greenwald et al., Nucl. Fusion, 28, 2199 (1988)NUFUAU0029-551510.1088/0029-5515/28/12/009] in case of unintentional high-to-low confinement transition.

Full conversion from Ohmic to runaway electron driven current via massive gas injection in the TCV tokamak

Abstract Full conversion from Ohmic to runaway electron (RE) driven current was observed in the TCV tokamak following massive injection of neon through a disruption mitigation valve into a low-density limited circular plasma. Following a partial disruption, a stable 200 kA RE beam is maintained for more than one second. Controlled ramp-down of the RE beam with adjustable decay rate was demonstrated. Control of the beam vertical position was achieved down to a RE current of 20 kA. RE beam formation is observed in elongated plasma configurations up to κ=1.5. A reproducible scenario for RE beam generation without loss of circulating current is of particular interest for disruption modelling applications.

Enhanced confinement in diverted negative-triangularity L-mode plasmas in TCV

The favorable confinement properties of negative-triangularity (NT) tokamak configurations were discovered in the TCV tokamak in the late 1990s and were documented over the two following decades, through investigations of predominantly electron-heated plasmas in limited topologies. The most recent experimental campaign in TCV has marked a leap forward, characterized by the development of a variety of diverted NT shapes that are robustly stable with basic Ohmic heating. The application of auxiliary heating, directed now at both electrons and ions (using electron-cyclotron resonance heating as well as neutral-beam injection), has enabled the achievement of record performances for L-mode plasmas, with normalized β values reaching 2.8 transiently (as well as 2 in steady state, but reverting to a limited configuration) and with comparable ion and electron temperatures. The systematic confinement enhancement with NT is confirmed in these experiments. The L-mode existence space is broader than at positive triangularity, with only sporadic transitions to H-mode observed up to 1.4 MW heating power regardless of the magnetic-field-gradient direction relative to the X-point. These experiments are planned to be continued with even higher power following a heating-source upgrade.

Cherenkov probes and runaway electrons diagnostics

The beams of fast runaway electrons (RE), which are often produced during tokamak discharges, are particularly dangerous and can induce serious damages of the vacuum vessel and internal components of the machine. The proper and fast diagnostics of RE beams is essential for controlling the discharge, e.g., by early mitigation of disruptions and potentially dangerous RE beams. The diagnostics of RE beams is usually based on measurements of the radiation emitted either by these electrons, or as a result of their interactions with plasma and/or vessel walls. Such a radiation is usually recorded by the means of probes placed outside the vacuum vessel. The method developed by our team is based on the probe located inside the vacuum vessel. The probe can be used to detect highly localized RE bunches and to determine their spatial and temporal characteristics. During last few years, the NCBJ team have developed and used the RE diagnostics based on the Cherenkov effect observed in diamond radiators coupled with fast photomultipliers. During the investigated discharges, the probe was inserted into the vacuum vessel, and its head was placed at the plasma edge, where fast RE are expected. A correlation between signals recorded using our probes and other diagnostics, e.g., hard x-ray signals, was also studied. In this paper, we present recent results of the RE measurements by means of Cherenkov probes, which were performed in the COMPASS and TCV tokamaks.

The role of plasma-molecule interactions on power and particle balance during detachment on the TCV tokamak

This paper shows experimental results from the TCV tokamak that indicate plasma-molecule interactions involving and possibly D− play an important role as sinks of energy (through hydrogenic radiation as well as dissociation) and particles (ions) during divertor detachment if low target temperatures (<3 eV) are achieved. Both molecular activated recombination (MAR) and ion source reduction due to a power limitation effect are shown to be important in reducing the ion target flux during a density ramp. In contrast, the electron–ion recombination (EIR) ion sink is too small to play an important role in reducing the ion target flux. MAR or EIR do not occur during N2 seeding induced detachment as the target temperatures are not sufficiently low. The impact of is shown to be underestimated in present (vibrationally unresolved) SOLPS-ITER simulations, which could result from an underestimated rate. The converged SOLPS-ITER simulations are post-processed with alternative reaction rates, resulting in considerable contributions of to particle and power losses as well as dissociation below the D2 dissociation area. Those findings are in quantitative agreement with the experimental results.

Integrated Real-Time Supervisory Management for Off-Normal-Event Handling and Feedback Control of Tokamak Plasmas

For long-pulse tokamaks, one of the main challenges in control strategy is to simultaneously reach multiple control objectives and to robustly handle in real-time (RT) unexpected events (off-normal-events -- ONEs) with a limited set of actuators. We have developed in our previous work a generic architecture of the plasma control system (PCS) including a supervisor and an actuator manager to deal with these issues. We present in this paper recent developments of real-time decision-making by the supervisor to switch between different control scenarios (normal, backup, shutdown, disruption mitigation, etc.) during the discharge, based on off-normal-event states. We first standardize the evaluation of ONEs and thereby simplify significantly the supervisor decision logic, as well as facilitate the modifications and extensions of ONE states in the future. The whole PCS has been implemented on the TCV tokamak, applied to disruption avoidance with density limit experiments, demonstrating the excellent capabilities of the new RT integrated strategy.

Millimeter-wave beam scattering and induced broadening by plasma turbulence in the TCV tokamak

The scattering of millimeter-wave beams from electron density fluctuations and the associated beam broadening are experimentally demonstrated. Using a dedicated setup, instantaneous deflection and (de-)focusing of the beam due to density blobs on the beam path are shown to agree with full-wave simulations. The detected time-averaged wave power transmitted through the turbulent plasma is reproduced by the radiative-transfer model implemented in the WKBeam code, which predicts a ∼50% turbulence-induced broadening of the beam cross-section. The role of core turbulence for the considered geometry is highlighted.

Suppression of first-wall interaction in negative triangularity plasmas on TCV

Magnetically confined fusion plasmas with negative triangularity (δ) exhibit greater L-mode confinement than with positive δ. Recent experiments in the TCV and DIII-D tokamaks have correlated the confinement improvement to a reduction of fluctuations within the plasma core. We report on fluctuation measurements in the scrape-off layer (SOL) for −0.61 < δ < +0.64 in limited and diverted ohmic L-mode plasmas; these reveal a strong reduction in SOL fluctuation amplitudes at δ ≲ −0.25, and, surprisingly, an almost full suppression of plasma interaction with the main-chamber first-wall, which could have important implications for the prospects of using negative δ plasmas as a reactor solution. An exploration of several physical mechanisms suggests that a reduced connection length—intrinsic to negative δ plasmas—plays a critical role in the origin of this phenomenon.

A study of the influence of plasma–molecule interactions on particle balance during detachment

In this work we provide experimental insights into the impact of plasma-molecule interactions on the target ion flux decrease during divertor detachment achieved through a core density ramp in the TCV tokamak. Our improved analysis of the hydrogen Balmer series shows that plasma-molecule processes are strongly contributing to the Balmer series intensities and substantially alter the divertor detachment particle balance. We find that Molecular Activated Recombination (MAR) ion sinks from $H_2^+$ and/or $H^-$ are a factor $\sim$ 5 larger than Electron-Ion Recombination (EIR) and are a significant contributor to the observed reduction in the outer divertor ion target flux. Molecular Activated Ionisation (MAI) may also be significant during detachment. Plasma-molecule interactions enhance the Balmer line series emission strongly near the target as detachment proceeds. This indicates enhancements of the Lyman series, potentially affecting power balance in the divertor. As those enhancements vary spatially in the divertor and are different for different transitions, they are expected to result in a separation of the $Ly\beta$ and $Ly\alpha$ emission regions. This may have implications for the treatment and diagnosis of divertor opacity. The demonstrated enhancement of the Balmer series through plasma-molecule processes potentially poses a challenge to using the Balmer series for understanding and diagnosing detachment based only on atom-plasma processes.

Non-evaporable getter pump operations in the TCV tokamak

A non-evaporable getter pump using the ZAO alloy was installed on TCV and operated during the 2019 experimental campaign. The pump performance, determined from in-situ pumping speed measurements, indicates that current getter pump technologies are well suited to tokamak operations with high power exhaust plasmas, and provide an interesting alternative for applications in fusion experiments.

H-mode scrape-off layer power width in the TCV tokamak

Obtaining acceptable conditions at the divertor targets of a next-step fusion experiment based on the tokamak concept is expected to be particularly challenging because of the small predicted value of the plasma power exhaust channel width. An increased confidence in this prediction is important to forestall any power exhaust issue and in developing corresponding divertor solutions. With the present prediction relying on empirical scaling laws based on data from six tokamaks, this letter tests these scaling laws on an additional device, the TCV tokamak. Estimates of the exhaust channel width, λ q , based on Thomson scattering measurements of the electron temperature and density profiles, correlate well with outer target infrared thermography. Reasonable agreement with multi-device scaling laws is found only when including both the power crossing the separatrix and the Greenwald density fraction as regression parameters. TCV’s λ q is 2 to 3 times smaller than in spherical tokamaks for the same value of the poloidal field. The inclusion of TCV data in the scaling laws would, therefore, require the retention of an explicit aspect ratio dependence, with consequences for all other dependencies.

Study of suprathermal electron dynamics during electron cyclotron current drive using hard x-ray measurements in the TCV tokamak

Suprathermal electron dynamics in real and momentum spaces plays an important role in the physics of electron cyclotron current drive (ECCD), in particular for predicting the wave driven current. In the TCV tokamak, a hard x-ray spectroscopy system equipped with digital pulse detection enables detailed studies of the creation and relaxation of suprathermal electrons. The time evolution of hard x-ray spatial profiles and energy spectra is observed utilizing a minimally perturbative ECCD modulation technique. Time-dependent Fokker-Planck modeling combined with a hard x-ray synthetic diagnostic is used to interpret experimental results and test various suprathermal electron transport models. The comparison suggests a dependency of the radial transport on the EC wave power.

Numerical investigation of optimal divertor gas baffle closure on TCV

A first set of divertor gas baffles has recently been installed in the TCV tokamak. In order to explore the physics determining the benefits and limitations of divertor baffling and to guide the design of a possible second generation of baffles, the effect of baffle closure is investigated using the 2D transport code SolEdge2D-EIRENE with realistic wall geometries. The baffle extension is scanned, first imposing the same upstream conditions as in previous SOLPS-ITER studies, then extending the parameter space to access detached plasma conditions. In attached plasma cases, divertor neutral compression is maximised by a Low-Field Side baffle length with an opening between the separatrix and the baffle tip of approximately 5 λ q , resulting in an increase in neutral compression by a factor 4 with respect to the unbaffled case. In detached cases this ratio can be improved by up to a factor 25 using higher baffle closures. This difference in behaviour between attached and detached conditions is explained by a model based on the ionisation mean free path of neutral particles recycled from the target. In some conditions, the optimal baffle extension in terms of neutral compression is found to be subject to high levels of intercepted upstream heat flux, which results in a peak heat flux on the baffles comparable to the one impinging on the outer target. The individual roles of the High-Field Side and Low-Field Side baffles are disentangled by means of dedicated simulations, which show a lower global impact of the inner baffle. This study suggests that an outer baffle with a gap of approximately 3 λ q , slightly more closed than the one presently installed, could further enhance the neutral compression ratio in cases where the ionisation front is detached. The biggest unknown in these simulations is related to far SOL particle transport, which could result in higher levels of baffle recycling and thus limit baffle performance.

Observation of Alfvén Eigenmodes driven by off-axis neutral beam injection in the TCV tokamak

Fast-particle driven Alfvén Eigenmodes have been observed in low-collisionality discharges with off-axis neutral beam injection (NBI), electron cyclotron resonance heating (ECRH) and a reduced toroidal magnetic field. During NBI and ECRH, toroidicity induced Alfvén Eigenmodes (TAEs) appear in frequency bands close to 200 kHz and energetic-particle-induced geodesic acoustic modes (EGAMs) are observed at about 40 and 80 kHz. When turning off ECRH in the experiment, those beam-driven modes disappear which can be explained by a modification of the fast-ion slowing down distribution. In contrast, coherent fluctuations close to the frequency of the beam-driven TAEs are present throughout the experiment. The modes are even observed during ohmic plasma conditions, which clearly demonstrates that they are not caused by fast particles and suggests an alternative drive, such as turbulence. The mode-induced fast-ion transport has been found to be weak and marginal in terms of the fast-ion diagnostic sensitivities. Measurements of the plasma stored energy, neutron rates, neutral particle fluxes and fast-ion D-alpha spectroscopy show good agreement with neoclassical modelling results from TRANSP. This is further supported by a similarly good agreement between measurement and modelling in cases with and without ECRH and therefore with and without the modes. Instead, a significant level of charge exchange losses are predicted and observed which generate a bump-on-tail fast-ion distribution function that can provide the necessary free energy to EGAMs.

Modeling of the response function of CdTe detectors for a hard X-ray synthetic diagnostic

Abstract The response function of CdTe detectors used in a hard X-ray spectrometer system (HXRS) on the TCV tokamak fusion experiment is modeled based on the measured energy spectra from multiple gamma-ray sources. The response function consists of an asymmetric Gaussian photopeak, an exponentially inclined shelf structure below the photopeak, and CdTe escape peaks. It is demonstrated that the measured energy spectra of radioactive sources can be correctly reconstructed using this response function. This model is applied to a hard X-ray synthetic diagnostic used to convert a modeled electron distribution function to a measured hard X-ray signal, for the study of energetic electrons in the TCV tokamak.

Classification of tokamak plasma confinement states with convolutional recurrent neural networks

During a tokamak discharge, the plasma can vary between different confinement regimes: low (L), high (H) and, in some cases, a temporary (intermediate state), called dithering (D). In addition, while the plasma is in H mode, edge localized modes (ELMs) can occur. The automatic detection of changes between these states, and of ELMs, is important for tokamak operation. Motivated by this, and by recent developments in deep learning, we developed and compared two methods for automatic detection of the occurrence of L-D-H transitions and ELMs, applied on data from the TCV tokamak. These methods consist in a convolutional neural network and a convolutional long short term memory neural network. We measured our results with regards to ELMs using ROC curves and Youden’s score index, and regarding state detection using Cohen’s Kappa index.

MANTIS: A real-time quantitative multispectral imaging system for fusion plasmas.

This work presents a novel, real-time capable, 10-channel Multispectral Advanced Narrowband Tokamak Imaging System installed on the TCV tokamak, MANTIS. Software and hardware requirements are presented together with the complete system architecture. The image quality of the system is assessed with emphasis on effects resulting from the narrowband interference filters. Some filters are found to create internal reflection images that are correlated with the filters' reflection coefficient. This was measured for selected filters where significant absorption (up to 65% within ∼70 nm of the filter center) was measured. The majority of this was attributed to the filter's design, and several filters' performance is compared. Tailored real-time algorithms exploiting the system's capabilities are presented together with benchmarks comparing polling and event based synchronization. The real-time performance is demonstrated with a density ramp discharge performed on TCV. The behavior of spectral lines' emission from different plasma species and their interpretation are qualitatively described.

Suprathermal electron driven fishbone instability in the TCV tokamak

Observations of fishbone oscillations driven by energetic electrons have been reported in several tokamaks, but experimental evidence of the interaction between electrons and the mode has remained limited. In the TCV tokamak, a state-of-the-art multi-channel hard x-ray spectroscopy system equipped with digital pulse detection allows the study of the redistribution of electron cyclotron current drive (ECCD)-created suprathermal electrons at high temporal resolution. For the first time, the response of hard x-ray profiles to the internal kink mode is observed at the frequency of the mode. This work demonstrates the role of suprathermal electrons in destabilizing the fishbone mode and in particular the interaction of trapped electrons with the mode. The mode is destabilized after the ramp-up of suprathermal electrons and then trapped suprathermal electrons preferentially interact with the mode and are expelled during the mode oscillation phase. The experimentally observed mode frequency and rotation direction are in good agreement with the solution of the linear fishbone dispersion relation, where the electron distribution function is modelled using a Fokker-Planck code coupled with a hard x-ray synthetic diagnostic. The frequency and stability predicted by the modeling also agree well with data from an ECCD power scan, as the varying suprathermal electron distribution affects the mode behavior.