Two-dimensional inference of divertor plasma characteristics: advancements to a multi-instrument Bayesian analysis system

Abstract An integrated data analysis system based on Bayesian inference has been developed for application to data from multiple diagnostics over the two-dimensional cross-section of tokamak divertors. The divertor multi-instrument Bayesian analysis system (D-MIBAS) has been tested on a synthetic dataset (including realistic experimental uncertainties) generated from SOLPS-ITER predictions of the MAST-U divertor. The resulting inference was within 6%, 5%, and 30% median absolute percentage error of the SOLPS-ITER-predicted electron temperature, electron density, and neutral atomic hydrogen density, respectively, across a two-dimensional poloidal cross-section of the MAST-U Super-X outer divertor. To accommodate molecular contributions to Balmer emission, an advanced emission model has been developed. This is shown to be crucial for inference accuracy. Our D-MIBAS system utilises a mesh aligned to poloidal magnetic flux-surfaces, throughout the divertor, with plasma parameters assigned to each mesh vertex and collectively considered in the inference. This allowed comprehensive forward models of multiple diagnostics and the inclusion of expected physics. This is shown to be important for inference precision when including molecular contributions to Balmer emission. These developments pave the way for accurate, two-dimensional electron temperature, electron density, and neutral atomic hydrogen density inferences for MAST-U divertor experimental data for the first time.