Elevating zero dimensional global scaling predictions to self-consistent theory-based simulations

Abstract

We have developed an innovative workflow, STEP-0D, within the OMFIT integrated modelling framework. Through systematic validation against the International Tokamak Physics Activity (ITPA) global H-mode confinement database, we demonstrated that STEP-0D, on average, predicts the energy confinement time with a mean relative error (MRE) of less than 19%. Moreover, this workflow showed promising potential in predicting plasmas for proposed fusion reactors such as ARC, EU-DEMO, and CFETR, indicating moderate H-factors between 0.9 and 1.2. STEP-0D allows theory-based prediction of tokamak scenarios, beginning with zero-dimensional (0D) quantities. The workflow initiates with the PRO-create module, generating physically consistent plasma profiles and equilibrium using the same 0D quantities as the IPB98(y,2) confinement scaling. This sets the starting point for the STEP (Stability, Transport, Equilibrium, and Pedestal) module, which further iterates between theory-based physics models of equilibrium, core transport, and pedestal to yield a self-consistent solution. Given these attributes, STEP-0D not only improves the accuracy of predicting plasma performance but also provides a path towards a novel fusion power plant (FPP) design workflow. When integrated with engineering and costing models within an optimization, this new approach could eliminate the iterative reconciliation between plasma models of varying fidelity. This potential for a more efficient design process underpins STEP-0D's significant contribution to future fusion power plant development.