Predictive modelling of edge transport phenomena in ELMy H-mode tokamak fusion plasmas
Johnny-Stefan Lönnroth
Abstract
Open-access reader
Johnny-Stefan Lönnroth
Abstract
Open-access reader
This thesis discusses a range of work dealing with edge plasma transport in magnetically confined fusion plasmas by means of predictive transport modelling, a technique in which qualitative predictions and explanations are sought by running transport codes equipped with models for plasma transport and other relevant phenomena. The focus is on high confinement mode (H-mode) tokamak plasmas, which feature improved performance thanks to the formation of an edge transport barrier. H-mode plasmas are generally characterized by the occurrence of edge localized modes (ELMs), periodic eruptions of particles and energy, which limit confinement and may turn out to be seriously damaging in future large tokamak reactors. \n The thesis introduces schemes and models for qualitative study of the ELM phenomenon in predictive transport modelling. It aims to shed new light on the dynamics of ELMs using these models. It tries, again making use of the ELM modelling schemes, to explain various experimental observations, notably the loss of performance and increased ELM frequency observed in experiments with enhanced toroidal magnetic field ripple and in situations with strong external neutral gas puffing. Finally, it also tries to establish more generally the potential effects of ripple-induced thermal ion losses on H-mode plasma performance and ELMs. \n It is demonstrated that the ELM modelling schemes introduced in the thesis can qualitatively reproduce the experimental dynamics of a number of ELM regimes. Using a theory-motivated ELM model based on a linear model of instability, the dynamics of combined ballooning-peeling mode ELMs is studied. It is shown that the ELMs are most often triggered by a ballooning mode instability, which renders the plasma peeling mode unstable and causes the ELM to continue in a peeling mode phase. Understanding the dynamics of ELMs will be a key issue when it comes to controlling and mitigating the ELMs in future large tokamaks. By means of integrated modelling, it is shown that an experimentally observed increase in the ELM frequency and deterioration of plasma confinement triggered by external neutral gas puffing might be due to a transition from the second to the first ballooning stable region, as expressed in magnetohydrodynamics (MHD). The result may have implications on the control of ELMs and performance in future tokamaks. Modest pedestal performance and benign ELMs observed in the presence of toroidal magnetic field ripple in dimensionless pedestal identity experiments between the JET and JT-6OU tokamaks are explained through predictive transport modelling as resulting from ripple-induced thermal ion losses, more precisely from non-diffusive (direct) losses. It is shown that ripple losses need not necessarily have a detrimental influence on performance, but that there is a trade-off between performance and benignity of ELMs. The results may have widely felt implications, because ITER, the next major facility on the way towards a commercial fusion reactor, is foreseen to operate with a non-negligible level of toroidal magnetic field ripple.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
This thesis discusses a range of work dealing with edge plasma transport in magnetically confined fusion plasmas by means of predictive transport modelling, a technique in which qualitative predictions and explanations are sought by running transport codes equipped with models for plasma transport and other relevant phenomena. The focus is on high confinement mode (H-mode) tokamak plasmas, which feature improved performance thanks to the formation of an edge transport barrier. H-mode plasmas are generally characterized by the occurrence of edge localized modes (ELMs), periodic eruptions of particles and energy, which limit confinement and may turn out to be seriously damaging in future large tokamak reactors. \n The thesis introduces schemes and models for qualitative study of the ELM phenomenon in predictive transport modelling. It aims to shed new light on the dynamics of ELMs using these models. It tries, again making use of the ELM modelling schemes, to explain various experimental observations, notably the loss of performance and increased ELM frequency observed in experiments with enhanced toroidal magnetic field ripple and in situations with strong external neutral gas puffing. Finally, it also tries to establish more generally the potential effects of ripple-induced thermal ion losses on H-mode plasma performance and ELMs. \n It is demonstrated that the ELM modelling schemes introduced in the thesis can qualitatively reproduce the experimental dynamics of a number of ELM regimes. Using a theory-motivated ELM model based on a linear model of instability, the dynamics of combined ballooning-peeling mode ELMs is studied. It is shown that the ELMs are most often triggered by a ballooning mode instability, which renders the plasma peeling mode unstable and causes the ELM to continue in a peeling mode phase. Understanding the dynamics of ELMs will be a key issue when it comes to controlling and mitigating the ELMs in future large tokamaks. By means of integrated modelling, it is shown that an experimentally observed increase in the ELM frequency and deterioration of plasma confinement triggered by external neutral gas puffing might be due to a transition from the second to the first ballooning stable region, as expressed in magnetohydrodynamics (MHD). The result may have implications on the control of ELMs and performance in future tokamaks. Modest pedestal performance and benign ELMs observed in the presence of toroidal magnetic field ripple in dimensionless pedestal identity experiments between the JET and JT-6OU tokamaks are explained through predictive transport modelling as resulting from ripple-induced thermal ion losses, more precisely from non-diffusive (direct) losses. It is shown that ripple losses need not necessarily have a detrimental influence on performance, but that there is a trade-off between performance and benignity of ELMs. The results may have widely felt implications, because ITER, the next major facility on the way towards a commercial fusion reactor, is foreseen to operate with a non-negligible level of toroidal magnetic field ripple.
Key concepts: Tokamak, Plasma, Magnetic confinement fusion, Ripple, Edge-localized mode, Mechanics, Toroid, Physics