부산대학교 도서관

Characterizing the dynamic behaviors of plasma properties in Hall-effect thruster discharge channels using traditional experimental techniques is challenging, as invasive probes perturb the plasma and optical techniques are limited by the presence of the channel walls. In this work, an alternative model-data fusion method is presented which allows for the time-resolved estimation of spatially averaged plasma properties extracted from discharge current signals. A physically constrained extended Kalman filter (EKF) is developed using a zero-dimensional (0D) global plasma model to estimate ion and neutral number densities, electron temperature, and electron bulk velocity. The filter is demonstrated to be able to robustly estimate plasma behavior over a wide range of operating conditions (i.e., varying discharge voltage and discharge current). The oscillatory behaviors of the estimated states are characterized based on amplitude and phase, and an analytical theory to describe the relative phases between the plasma properties and discharge current is developed. The technique is further extended to provide time-varying estimates of the anomalous electron scattering frequency for cross-field electron transport. It is found that all estimated states, as well as the frequencies and mobility, vary in dynamic behavior with thruster operating conditions and generally exhibit strong oscillations at the dominant frequency of the discharge current. The results from this work provide strong evidence that similar data-integrated models using additional measurements and more complete descriptions of the underlying plasma physics can provide insight into plasma behaviors that are traditionally challenging to measure.