Description
In this study, we develop a method for estimating edge plasma density profiles and analyzing beam attenuation using the KSTAR hydrogen beam emission spectroscopy (H-BES) system. The system, comprising 16 radial and 4 poloidal channels with approximately 1 cm spatial resolution, measures Doppler-shifted D-alpha emissions that serve as indicators of local plasma conditions. A collisional radiative model is employed to relate the observed D-alpha intensities to plasma density, with the beam’s influence accounted for by incorporating independently measured density and temperature profiles. We model the density profile using a Gaussian process prior and employ Markov Chain Monte Carlo (MCMC) methods to explore the posterior distribution. Beam attenuation, as observed through BES measurements, is used to refine the density estimation further. Importantly, the BES signals enable cross-validation between the inferred density profile and the measured beam attenuation, thereby providing mutual verification of these diagnostic parameters.
An absolute calibration factor is determined by integrating the BES data with edge interferometry measurements. In addition, our approach rigorously quantifies the uncertainties inherent in both the BES light intensity and the calibration process. This integrated treatment of uncertainties, along with the simultaneous analysis of density and beam attenuation, offers a more comprehensive understanding of beam–plasma interaction dynamics. By directly incorporating beam–plasma interactions into the model, our approach facilitates the determination of heating beam properties rather than relying solely on density profile-based estimations. We anticipate that further in-depth research will yield more detailed insights into beam properties during plasma operation.
