We present an ultrafast neural network (NN) model, QLKNN, which predicts core tokamak transport heat and particle fluxes. QLKNN is a surrogate model based on a database of 300 million flux calculations of the quasilinear gyrokinetic transport model QuaLiKiz. The database covers a wide range of realistic tokamak core parameters. Physical features such as the existence of a critical gradient for the onset of turbulent transport were integrated into the neural network training methodology. We have coupled QLKNN to the tokamak modelling framework JINTRAC and rapid control-oriented tokamak transport solver RAPTOR. The coupled frameworks are demonstrated and validated through application to three JET shots covering a representative spread of H-mode operating space, predicting turbulent transport of energy and particles in the plasma core. JINTRAC-QLKNN and RAPTOR-QLKNN are able to accurately reproduce JINTRAC-QuaLiKiz T i,e and n e profiles, but 3 to 5 orders of magnitude faster. Simulations which take hours are reduced down to only a few tens of seconds. The discrepancy in the final source-driven predicted profiles between QLKNN and QuaLiKiz is on the order 1%-15%. Also the dynamic behaviour was well captured by QLKNN, with differences of only 4%-10% compared to JINTRAC-QuaLiKiz observed at mid-radius, for a study of density buildup following the L-H transition. Deployment of neural network surrogate models in multi-physics integrated tokamak modelling is a promising route towards enabling accurate and fast tokamak scenario optimization, Uncertainty Quantification, and control applications.

Fusion reactions of solid density boron-11 with protons after initiation of a fusion flame by very powerful picosecond laser pulses were derived for plane geometry. The problem of lateral energy losses with laser beams was solved by using spherical geometry, where however the gains are limited. The other elimination of losses now available by cylinderaxis symmetric 10 kilotesla magnetic fields is possible needing laser powers in the exawatt range. Estimations are presented by varying parameters for reducing the necessary laser pulse powers to lower values by up to a factor 100.

Integrating the plasma core performance with an edge and scrape-off layer (SOL) that leads to tolerable heat and particle loads on the wall is a major challenge. The new European medium size tokamak task force (EU-MST) coordinates research on ASDEX Upgrade (AUG), MAST and TCV. This multi-machine approach within EU-MST, covering a wide parameter range, is instrumental to progress in the field, as ITER and DEMO core/pedestal and SOL parameters are not achievable simultaneously in present day devices. A two prong approach is adopted. On the one hand, scenarios with tolerable transient heat and particle loads, including active edge localised mode (ELM) control are developed. On the other hand, divertor solutions including advanced magnetic configurations are studied. Considerable progress has been made on both approaches, in particular in the fields of: ELM control with resonant magnetic perturbations (RMP), small ELM regimes, detachment onset and control, as well as filamentary scrape-off-layer transport. For example full ELM suppression has now been achieved on AUG at low collisionality with n  =  2 RMP maintaining good confinement . Advances have been made with respect to detachment onset and control. Studies in advanced divertor configurations (Snowflake, Super-X and X-point target divertor) shed new light on SOL physics. Cross field filamentary transport has been characterised in a wide parameter regime on AUG, MAST and TCV progressing the theoretical and experimental understanding crucial for predicting first wall loads in ITER and DEMO. Conditions in the SOL also play a crucial role for ELM stability and access to small ELM regimes.

The time evolution of radiating vapour shields over eroding solid surfaces and the resulting erosion rates are modelled by one dimensional (1-D) and 1½-D resistive MHD codes. Graphite or carbonized divertor plates subjected to high energy deuterium plasma particles during disruptions or giant ELMs are considered. The energy flux range assumed corresponds to ITER conditions. Various physical phenomena having a primary effect on the erosion rate, such as collisional interaction of the energy carriers with the target (solid surface or vapour particles), electrostatic shielding, magnetohydrodynamic interaction and radiant energy transport, are investigated in detail. In the 1-D and 1½-D approximations used and for the energy input parameter range considered (Q0 = 1011 W/m2) ablation rates of the order of 1028 m-2·s-1 were obtained. It is shown that processes, such as lateral expansion, lateral drift, radiation losses through the lateral surfaces of the scrape-off layer (SOL) and the Hall effect, may notably change the predicted erosion rates and warrant a more elaborate, at least 2-D, treatment of the problem.

As a result of plasma-wall contact during hard disruptions, intense vaporization of the divertor plates is initiated. The evolving vapour layer intercepts a fraction of the incident energy carriers, thus shielding the plates. The plasma particles of different energies and masses moving along the skewed magnetic field lines penetrate the vapour layer to different depths, thus causing charge separation there and the onset of an electrostatic field. It is shown in a one dimensional approximation that, owing to the constraints imposed on the electric field and on the current components at the plate surface, a rather intense lateral drift motion of the vapour evolves in the scrape-off layer (SOL). The analysis shows that drift velocities of the order of 103 to 104 m/s are to be expected, which may notably impair the shielding characteristics of the vapour layers and increase the erosion rates of the divertor plates