Search

Two different approaches for predicting plasma toroidal velocity (v ) are developed and used in self-consistent simulations of H-mode plasmas with the presence of ITB using BALDUR integrated predictive modelling code. In the first approach, the toroidal velocity depends on the plasma current density; while in the second approach the toroidal velocity is directly proportional to the ion temperature. The profile of v is used to calculate the ExB flow shear which is a main mechanism for plasma transport suppression, leading to the ITB formation. In all simulations, the core transport model is a combination of NCLASS neoclassical transport and semi-empirical Mixed Bohm/gyro-Bohm model that includes the ITB effects. The boundary condition is set at top of the pedestal and is estimated using a pedestal model based on a combination of magnetic and flow shear stabilization pedestal width scaling and an infinite-n ballooning pressure gradient. Two toroidal velocity models are used to simulate the time evolution of plasma temperature and density profiles of 10 JET discharges. The root mean square error (RMSE) is used to compare simulation results of those 10 JET discharges with experimental data. It is found that RMSE of Ti , Te , ne are 28.1%, 31.8%, and 15.0% for the first toroidal velocity model and 25.5%, 30.2%, and 15.1% for the second toroidal velocity model, respectively. Furthermore, this suite of codes is used to predict the ITER performance under standard type I ELMy H-mode. It is found that the simulation yields formation of a narrow ITB near r/a = 0.7 in the simulation using the current density dependent model and a wide ITB from r/a = 0.5 to 0.8 in the simulation using the ion temperature dependent model. The average of central ion temperature, total fusion power output and alpha power are predicted to be 36 keV, 159 MW and 492 MW for the current density dependent model and 49 keV, 218 MW and 786 MW for the ion temperature dependent model, respectively.

Core-SOL simulations were carried out of plasma in tokamak reactors operating in a low confinement mode (L-mode), for various conditions that match available experimental data. The simulation results were quantitatively compared against experimental data, showing that the average RMS errors for electron temperature, ion temperature, and electron density were lower than 16% or less for 14 L-mode discharges from two tokamaks named DIII-D and TFTR. In the simulations, the core plasma transport was described using a combination of neoclassical transport calculated by NCLASS module and anomalous transport by Multi-Mode-Model version 2001 (MMM2001). The scrape-off-layer (SOL) is the small amount of residual plasma that interacts with the tokamak vessel, and was simulated by integrating the fluid equations, including sources, along open field lines. The SOL solution provided the boundary conditions of core plasma region on low confinement mode (L-mode). The experimental data were for 14 L-mode discharges and from two tokamaks, named DIII-D and TFTR.

It is well known that edge plasma parameters influence performance in many different ways (profile stiffness is probably one of the best known examples). In ELMy H-mode a thin region with improved transport characteristics (Edge Transport Barrier) co

The development of a predictive capability for the characteristics of the H-mode pedestal and ELMs is important in predicting tokamak reactor performance. The H-mode pedestal effects core confinement through temperature profile stiffness. The ELM power flux can significantly impact divertor design. We discuss progress in these areas through inter-machine comparison under the venue of the International Tokamak Physics Activity, ITPA. The pressure in the H-mode transport barrier region is typically set by the ELM cycle rather than transport. Several features of the ELM onset conditions are consistent with an ideal peeling-ballooning mode instability: 1) strong shape dependence, 2) mode structure, 3) reduced p' at high collisionality, 4) p' dependence on transport barrier width. Type I ELM energy loss is found to be proportional to the energy in the H-mode pedestal. There is some consistency between the peeling-ballooning mode radial width and the ELM size. Type I ELM energy loss is reduced at high density to levels that would be tolerable in a reactor scale device through reduced conductive loss and possibly shift to a more radially localized instability.