Plasma edge simulations including realistic wall geometry with SOLPS-ITER

A new hybrid fluid-kinetic approach for the hydrogenic neutrals (atoms and molecules) in the plasma edge is presented. The hybrid approach combines a fully kinetic model for the atoms in the low-collisional regions near the vessel wall, and for the molecules in the whole plasma edge domain, with a micro-macro approach for atoms originating from recycling at the divertor targets, volumetric recombination, and dissociation of molecules. With the micro-macro approach, the originally scattering-dominated collision term due to charge-exchange collisions in the kinetic equation is transformed to an absorption-dominated term, while a large part of the neutral population is treated through a fluid approach. For JET L-mode plasmas, the premature termination of Monte Carlo particle trajectories in the hybrid approach leads to a reduction of the CPU time by approximately a factor 3 for a high-recycling case and by approximately a factor 11 for a partially detached case compared with a simulation with fully kinetic neutrals and the same amount of particles. For coupled fluid plasma -hybrid neutral simulations -the hybrid approach predicts the plasma divertor target profiles with a maximum hybrid-kinetic discrepancy of approximately 30%.

Section: From Kinetic To Combined Spatially–micro–macro Hybrid Neutra...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Combination of micro–macro and spatially hybrid fluid‐kinetic approach for hydrogenic plasma edge neutrals

Horsten

Groth

Dekeyser

et al. 2022

“…The SOLPS‐ITER version employed in this study features the recently implemented unstructured solver, which allows flexible vessel geometry description and grids extending up to the first wall. [ 29 ] This version inherently accounts for grid misalignment, an important requirement to obtain reliable fluid neutral simulations. [ 30 ] Furthermore, the most up‐to‐date advanced fluid neutral models available [ 31 ] are employed here, which provide similar accuracy as computationally expensive Monte Carlo kinetic neutrals in highly collisional regimes, without the need for additional parameters in contrast to other fluid models, and avoiding statistical noise, which hampers gradient accuracy.…”

Section: Parameter Estimation Case Studymentioning

confidence: 99%

Bayesian maximum a posteriori‐estimation of κ turbulence model parameters using algorithmic differentiation in SOLPS‐ITER

Carli

Dekeyser

Blommaert

et al. 2021

Radial anomalous diffusion coefficients are among the largest sources of uncertainty in mean-field plasma-edge codes such as SOLPS-ITER. These coefficients are machine, scenario, and space-dependent, thus hampering the predictive and interpretive capability of these codes. In fact, modellers usually adjust these coefficients manually, based on expert judgement or on large, computationally expensive, parameter scans. Matching data from various diagnostics, each with their own experimental uncertainties, additionally complicates the problem of finding a good parameter set. In addition, standard non-linear regression techniques have shown to become prohibitive for expensive plasma-edge simulations with many unknown parameters, as gradient calculation was based on finite differences. In this paper, we apply algorithmic differentiation (AD) as an efficient and more accurate alternative for gradient calculation in large, continuously developed codes as SOLPS-ITER. In addition, the lack of uncertainty information and danger of data overfitting, key limitations of regression techniques, are overcome by combining data and uncertainties from different diagnostics in a Bayesian framework. We implement for the first time such a Bayesian inference framework into SOLPS-ITER, using gradient-based optimization with gradients obtained through tangent AD to find the maximum a posteriori (MAP) values of the parameters. The recently developed κ turbulence model is employed to limit the number of unknown parameters compared to full spatial profiles of diffusion coefficients. The Bayesian MAP-estimation is compared to standard regression techniques on a small-scale tokamak. We adopt fictitious experimental data obtained from a reference SOLPS-ITER solution with artificial measurement noise.

“…Options to include the full chamber into the computational domain were carried out in previous SOLPS versions 4 and 5.0, [2][3][4] and active development is on-going to include this capability into the latest SOLPS-ITER version. [5] Many reports have been made on the formation of density shoulders in the SOL under high divertor density regimes. [6] In such conditions, which are relevant for ITER, plasma is strongly transported to the main wall in the far-SOL, increasing the role of plasma-wall interactions there.…”

Section: Introductionmentioning

confidence: 99%

“…Options to include the full chamber into the computational domain were carried out in previous SOLPS versions 4 and 5.0, [ 2–4 ] and active development is on‐going to include this capability into the latest SOLPS‐ITER version. [ 5 ]…”

Section: Introductionmentioning

confidence: 99%

SOLEDGE3X full vessel plasma simulations for computation of ITER first‐wall fluxes

Rivals

Tamain

Marandet

et al. 2022

ITER Pre‐Fusion Power Operation 1 (PFPO‐1) phase low‐power plasmas are simulated with the SOLEDGE3X‐EIRENE code, a multi‐fluid edge plasma solver that can describe plasma conditions up to the machine's first wall, including a kinetic description of neutrals. Here, SOLEDGE3X is used in 2D mean‐field mode. A throughput scan is performed, reproducing a similar scan from the ITER SOLPS simulations database. The physics assumptions for these simulations are presented, in particular the plasma–neutral interaction model where molecule charge‐exchange and ion‐molecule elastic collisions have newly been added. Challenges encountered in such up‐to‐the‐wall simulations are discussed, along with remaining open questions. Results are presented with two foci: on targets, and on the rest of the first wall. On targets, conditions transition from attached to partially detached as expected, and toroidally symmetric heat fluxes (i.e. without 3D effects) remain below 5 MW/m2, within the divertor design limits. An increase in the parallel heat flux decay length λq with throughput is also observed. On the rest of the first wall, increasing puff rate does not only spread power deposited on the target but also increases heat load on the first wall. These fluxes remain however small (<100 kW/m2).