-
Neural-Parareal: Dynamically Training Neural Operators as Coarse Solvers for Time-Parallelisation of Fusion MHD Simulations
Authors:
S. J. P. Pamela,
N. Carey,
J. Brandstetter,
R. Akers,
L. Zanisi,
J. Buchanan,
V. Gopakumar,
M. Hoelzl,
G. Huijsmans,
K. Pentland,
T. James,
G. Antonucci,
the JOREK Team
Abstract:
The fusion research facility ITER is currently being assembled to demonstrate that fusion can be used for industrial energy production, while several other programmes across the world are also moving forward, such as EU-DEMO, CFETR, SPARC and STEP. The high engineering complexity of a tokamak makes it an extremely challenging device to optimise, and test-based optimisation would be too slow and to…
▽ More
The fusion research facility ITER is currently being assembled to demonstrate that fusion can be used for industrial energy production, while several other programmes across the world are also moving forward, such as EU-DEMO, CFETR, SPARC and STEP. The high engineering complexity of a tokamak makes it an extremely challenging device to optimise, and test-based optimisation would be too slow and too costly. Instead, digital design and optimisation must be favored, which requires strongly-coupled suites of High-Performance Computing calculations. In this context, having surrogate models to provide quick estimates with uncertainty quantification is essential to explore and optimise new design options. Furthermore, these surrogates can in turn be used to accelerate simulations in the first place. This is the case of Parareal, a time-parallelisation method that can speed-up large HPC simulations, where the coarse-solver can be replaced by a surrogate. A novel framework, Neural-Parareal, is developed to integrate the training of neural operators dynamically as more data becomes available. For a given input-parameter domain, as more simulations are being run with Parareal, the large amount of data generated by the algorithm is used to train new surrogate models to be used as coarse-solvers for future Parareal simulations, leading to progressively more accurate coarse-solvers, and thus higher speed-up. It is found that such neural network surrogates can be much more effective than traditional coarse-solver in providing a speed-up with Parareal. This study is a demonstration of the convergence of HPC and AI which simply has to become common practice in the world of digital engineering design.
△ Less
Submitted 2 May, 2024;
originally announced May 2024.
-
MHD simulations of small ELMs at low triangularity in ASDEX Upgrade
Authors:
A. Cathey,
M. Hoelzl,
G. Harrer,
M. G. Dunne,
G. T. A. Huijsmans,
K. Lackner,
S. J. P. Pamela,
E. Wolfrum,
S. Günter,
the JOREK team,
the ASDEX Upgrade Team,
the EUROfusion MST1 Team
Abstract:
The development of small- and no-ELM regimes for ITER is a high priority topic due to the risks associated to type-I ELMs. By considering non-linear extended MHD simulations of the ASDEX Upgrade tokamak with the JOREK code, we probe a regime that avoids type-I ELMs completely provided that the separatrix density is high enough. The dynamics of the pedestal in this regime are observed to be qualita…
▽ More
The development of small- and no-ELM regimes for ITER is a high priority topic due to the risks associated to type-I ELMs. By considering non-linear extended MHD simulations of the ASDEX Upgrade tokamak with the JOREK code, we probe a regime that avoids type-I ELMs completely provided that the separatrix density is high enough. The dynamics of the pedestal in this regime are observed to be qualitatively similar to the so-called quasi-continuous exhaust (QCE) regime in several ways. Repetitive type-I ELMs are substituted by roughly constant levels of outwards transport caused by peeling-ballooning modes (with dominant ballooning characteristics) which are localised in the last 5\% of the confined region (in normalised poloidal flux). The simulated low triangularity plasma transitions to a type-I ELMy H-mode if the separatrix density is sufficiently reduced or if the input heating power is sufficiently increased. The stabilising factors that play a role in the suppression of the small ELMs are also investigated by analysing the simulations, and the importance of including diamagnetic effects in the simulations is highlighted. By considering a scan in the pedestal resistivity and by measuring the poloidal velocity of the modes (and comparing to theoretical estimates for ideal and resistive modes), we identify the underlying instabilities as resistive peeling-ballooning modes. Decreasing the resistivity below experimentally-relevant conditions (i.e., going towards ideal MHD), the peeling-ballooning modes that constrain the pedestal below the type-I ELM stability boundary display sharply decreasing growth rates.
△ Less
Submitted 15 October, 2021;
originally announced October 2021.
-
Comparing spontaneous and pellet-triggered ELMs via non-linear extended MHD simulations
Authors:
A. Cathey,
M. Hoelzl,
S. Futatani,
P. T. Lang,
K. Lackner,
G. T. A. Huijsmans,
S. J. P. Pamela,
S. Günter,
the JOREK team,
the ASDEX Upgrade Team,
the EUROfusion MST1 Team
Abstract:
Injecting frozen deuterium pellets into an ELMy H-mode plasma is a well established scheme for triggering edge localized modes (ELMs) before they naturally occur. Based on an ASDEX Upgrade H-mode plasma, this article presents a comparison of extended MHD simulations of spontaneous type-I ELMs and pellet-triggered ELMs allowing to study their non-linear dynamics in detail. In particular, pellet-tri…
▽ More
Injecting frozen deuterium pellets into an ELMy H-mode plasma is a well established scheme for triggering edge localized modes (ELMs) before they naturally occur. Based on an ASDEX Upgrade H-mode plasma, this article presents a comparison of extended MHD simulations of spontaneous type-I ELMs and pellet-triggered ELMs allowing to study their non-linear dynamics in detail. In particular, pellet-triggered ELMs are simulated by injecting deuterium pellets into different time points during the pedestal build-up described in [A. Cathey et al. Nuclear Fusion 60, 124007 (2020)]. Realistic ExB and diamagnetic background plasma flows as well as the time dependent bootstrap current evolution are included during the build-up to capture the balance between stabilising and destabilising terms for the edge instabilities accurately. Dependencies on the pellet size and injection times are studied. The spatio-temporal structures of the modes and the resulting divertor heat fluxes are compared in detail between spontaneous and triggered ELMs. We observe that the premature excitation of ELMs by means of pellet injection is caused by a helical perturbation described by a toroidal mode number of n = 1. In accordance with experimental observations, the pellet-triggered ELMs show reduced thermal energy losses and narrower divertor wetted area with respect to spontaneous ELMs. The peak divertor energy fluency is seen to decrease when ELMs are triggered by pellets injected earlier during the pedestal build-up.
△ Less
Submitted 11 February, 2021;
originally announced February 2021.
-
Non-linear extended MHD simulations of type-I edge localised mode cycles in ASDEX Upgrade and their underlying triggering mechanism
Authors:
Andres Cathey,
M. Hoelzl,
K. Lackner,
G. T. A. Huijsmans,
M. G. Dunne,
E. Wolfrum,
S. J. P. Pamela,
F. Orain,
S. Günter,
the JOREK team,
the ASDEX Upgrade Team,
the EUROfusion MST1 Team
Abstract:
A triggering mechanism responsible for the explosive onset of edge localised modes (ELMs) in fusion plasmas is identified by performing, for the first time, non-linear magnetohydrodynamic simulations of repetitive type-I ELMs. Briefly prior to the ELM crash, destabilising and stabilising terms are affected at different timescales by an increasingly ergodic magnetic field caused by non-linear inter…
▽ More
A triggering mechanism responsible for the explosive onset of edge localised modes (ELMs) in fusion plasmas is identified by performing, for the first time, non-linear magnetohydrodynamic simulations of repetitive type-I ELMs. Briefly prior to the ELM crash, destabilising and stabilising terms are affected at different timescales by an increasingly ergodic magnetic field caused by non-linear interactions between the axisymmetric background plasma and growing non-axisymmetric perturbations. The separation of timescales prompts the explosive, i.e. faster than exponential, growth of an ELM crash which lasts ${\sim}$ 500 $μ$s. The duration and size of the simulated ELM crashes compare qualitatively well with type-I ELMs in ASDEX Upgrade. As expected for type-I ELMs, a direct proportionality between the heating power in the simulations and the ELM repetition frequency is obtained. The simulations presented here are a major step forward towards predictive modelling of ELMs and of the assessment of mitigation techniques in ITER and other future tokamaks.
△ Less
Submitted 26 October, 2020; v1 submitted 20 July, 2020;
originally announced July 2020.
-
Electron kinetics inferred from observations of microwave bursts during edge localised modes in the Mega-Amp Spherical Tokamak
Authors:
S. J. Freethy,
K. G. McClements,
S. C. Chapman,
R. O. Dendy,
W. N. Lai,
S. J. P. Pamela,
V. F. Shevchenko,
R. G. L. Vann
Abstract:
Recent measurements of microwave and X-ray emission during edge localised mode (ELM) activity in tokamak plasmas provide a fresh perspective on ELM physics. It is evident that electron kinetics, which are not incorporated in standard (fluid) models for the instability that drives ELMs, play a key role in the new observations. These effects should be included in future models for ELMs and the ELM c…
▽ More
Recent measurements of microwave and X-ray emission during edge localised mode (ELM) activity in tokamak plasmas provide a fresh perspective on ELM physics. It is evident that electron kinetics, which are not incorporated in standard (fluid) models for the instability that drives ELMs, play a key role in the new observations. These effects should be included in future models for ELMs and the ELM cycle. The observed radiative effects paradoxically imply acceleration of electrons parallel to the magnetic field combined with rapid acquisition of perpendicular momentum. It is shown that this paradox can be resolved by the action of the anomalous Doppler instability which enables fast collective radiative relaxation, in the perpendicular direction, of electrons accelerated in the parallel direction by inductive electric fields generated by the initial ELM instability.
△ Less
Submitted 4 November, 2014;
originally announced November 2014.
