Pedestal electron collisionality and toroidal rotation during ELM-crash suppression phase under n = 1 RMP in KSTAR

Self Cite

KSTAR has clarified a set of unresolved 3-D physics issues utilizing the ITER-like in-vessel, three-row, resonant magnetic perturbation (RMP) configurations. Since RMP-driven, edge-localized-modes (ELM)-crash control elevates the divertor heat flux peak through its impact on edge plasma parameters and transport, a series of intentionally misaligned RMP configurations (IMC) have been explored to investigate the relationship between RMP ELM control and divertor heat fluxes, while searching for an ideal IMC that could be favorable in both aspects. First of all, the contrasting influence of kink vs anti-kink phasing on the ELM-crash suppression has been articulated, demonstrating the synergistic benefit of ‘kink’ phasing on ELM-crash-suppression. On the other hand, the three-row IMC in the anti-kink phasing becomes more insensitive to the ELM-crashes at the sub-marginal level of RMP, consistent with theory. Meanwhile, the divertor ‘wetted’ area of ELM-crash-suppression gets narrower than that of ELM-crash-mitigation, suggesting that ELM-crash-mitigation remains advantageous over ELM-crash-suppression in terms of time-averaged divertor thermal loading. In comparison, based on a set of two-row IMCs, no evidence of divertor heat flux broadening was found during ELM-crash-suppression, supporting a hypothesis that the dispersal of the divertor heat flux in 3-row IMCs cannot be driven by helically structured 2-row RMPs alone. Among ITER-like three-rows, lower two-row RMPs have been found to be much more effective in suppressing the ELM-crashes than upper two-row RMPs. Although it is quite preliminary, the up/down asymmetric dependence of RMP coupling may be generically attributed to lower-single-null (LSN) plasmas. Such a holistic understanding of RMP-driven, ELM-crash-control in KSTAR is expected not only to elucidate various subtle points in the vicinity of ELM-crash-suppression, but also to clarify the relevant divertor thermal loading issues for ITER and beyond.

Section: Discussionmentioning

confidence: 99%

Toward holistic understanding of the ITER-like resonant magnetic perturbation (RMP) ELM control on KSTAR

In¹,

Lee²,

Park³

et al. 2022

Self Cite

“…The discharge takes the shape of a diverted lower single null with moderate elongation κ ≈ 1.7 and a high triangularity δ = (δ u + δ l )/2 ≈ 0.58: the target position of the bottom X point (R x = 1.38 m, Z x = −0.88 m) is known to be suitable for the ELM crash suppression and is fully established after t = 3.5 s. To minimize and exclude transient shaping effects, we set the auxiliary heating to occur after t = 3.5 s so that the first L-H transition occurs near the fully established shaping. To compare with the conventional feed-forward cases reported by KSTAR [17,29], the RMP configuration was set to n = 1 with 90 • phasing. In addition, q 95 is set around 4.8 right after the H-mode transition time at t = 3.14 s. When the first ELM crash occurs, q 95 ≈ 4.9. q 95 for the reference discharge at the H-mode transition moment is near the lower bound for the n = 1 ELM crash suppression [29,30].…”

Section: Setup For the Reference H-mode Dischargementioning

confidence: 99%

“…To compare with the conventional feed-forward cases reported by KSTAR [17,29], the RMP configuration was set to n = 1 with 90 • phasing. In addition, q 95 is set around 4.8 right after the H-mode transition time at t = 3.14 s. When the first ELM crash occurs, q 95 ≈ 4.9. q 95 for the reference discharge at the H-mode transition moment is near the lower bound for the n = 1 ELM crash suppression [29,30].…”

Section: Setup For the Reference H-mode Dischargementioning

confidence: 99%

Preemptive RMP-driven ELM crash suppression automated by a real-time machine-learning classifier in KSTAR

Shin¹,

Hahn²,

Kim³

et al. 2022

Self Cite

Suppression or mitigation of edge-localized mode (ELM) crashes is necessary for ITER. The strategy to suppress all the ELM crashes by the resonant magnetic perturbation (RMP) should be applied as soon as the first low-to-high confinement (L-H) transition occurs. A control algorithm based on real-time machine learning (ML) enables such an approach: it classifies the H-mode transition and the ELMy phase in real-time and automatically applies the preemptive RMP. This paper reports the algorithm design, which is now implemented in the KSTAR plasma-control system, and the corresponding experimental demonstration of typical high-δ KSTAR H-mode plasmas. As a result, all initial ELM crashes are suppressed with an acceptable safety factor at the edge (q95) and with RMP field adjustment. Moreover, the ML-driven ELM-crash-suppression discharges remain stable without further degradation due to the regularization of the plasma pedestal.

“…Ideally, full suppression of ELMs is desirable, but access and sustainment of ELM-suppressed scenarios in reactor-scale devices with detached divertor conditions is uncertain, particularly in a device with limited torque injection available relative to the plasma inertia. Suppression of ELMs using applied 3D resonant magnetic perturbations (RMPs) has been demonstrated in many tokamak devices [7][8][9][10]. In DIII-D, RMP ELM suppression appears to require a rotation profile where the E × B rotation crosses zero near the top of the pedestal, which has thus far not been accessible with zero net injected torque [11,12].…”

Section: Introductionmentioning

confidence: 99%

Pellet triggering of edge localized modes in low collisionality pedestals at DIII-D

Wilcox¹,

Baylor²,

Bortolon³

et al. 2021

Edge localized modes (ELMs) are triggered using deuterium pellets injected into plasmas with ITER-relevant low collisionality pedestals, and the resulting peak ELM energy fluence is reduced by approximately 25-50% relative to natural ELMs destabilized at similar pedestal pressures. Cryogenically frozen deuterium pellets are injected from the low-field side of the DIII-D tokamak at frequencies lower than the natural ELM frequency, and heat flux is measured by infrared cameras. Ideal MHD pedestal stability calculations show that without pellet injection, these low collisionality pedestals were limited by their current density (peeling-limited) rather than their pressure gradient (ballooning-limited). ELM triggering success correlates strongly with pellet mass, consistent with the theory that a large pressure perturbation is required to trigger an ELM in low collisionality discharges that are far from the ballooning stability boundary. For sufficiently large pellets, both instantaneous and time-integrated ELM energy deposition measured by infrared cameras is reduced with respect to naturally occurring ELMs at the inner strike point, which is the position where it is largest for natural ELMs. Energy fluence at the outer strike point is less effected. Cameras observing both heat flux and D-alpha emission often find significant toroidally asymmetric striations in the outboard far scrape-off layer resulting from ELMs that are triggered by pellets. Toroidal asymmetries at the inner strike point are similar between natural and pellet-triggered ELMs, suggesting that the reduction in peak heat flux and total fluence at that location is robust for the conditions reported here.