Theory based prediction of tokamak energy confinement scaling

A new tool is developed for quick estimation of the energy confinement in a tokamak based on flux matching TGLF at a single mid-radius point (r/a=0.6) and EPED neural network. This model addresses one of the present limitations of systems studies, which currently rely on the commonly used experimental scaling law τ98,y2 to project plasma energy confinement. A benchmark of this model is performed against full transport simulations, and good agreement is found under reactor conditions and when the core scale length does not change significantly across the core plasma. A web-app is developed using this model for easy exploration of energy confinement. The model is compared to the τ98 dataset, where good agreement is seen. By systematically perturbing the global parameters of various existing and future tokamaks, an approximate theory based scaling law is developed. A linear regression of the simulation data shows significant differences in several global parameter scalings compared to τ98,y2 including major radius, aspect ratio, plasma current and elongation. The difference between the two scaling is currently being investigated.