Machine Learning Analysis of Hall Thruster Facility Effects Data – Penn State Behrend


Penn State Behrend


Cobalt Class (2019 – 2020)


Alec William Dady, Computer Science
Daniel Carter Donley, Software Engineering
James (Jimmy) Patrick Fennelly Jr., Software Engineering


Dr. Jason Frieman, NASA Glenn Research Center


Dr. Wen-Li Wang, Associate Professor of Computer Science and Software Engineering, Penn State Behrend


Terrestrial space simulation facilities cannot perfectly replicate the pressures observed in space. Since Hall thrusters are sensitive to the ambient pressure at which they are operated, this can lead to differences in performance between ground test and orbital operation. Despite extensive research, current physics-based models have not yet uncovered the cause of this sensitivity. Student teams are applying machine learning and/or data mining techniques to uncover previously undiscovered correlations in published data sets and devising methods to predict and/or correct for this sensitivity. The end result will include a report detailing the results of project including, but not limited to, training data, any models/analyses generated by the machine learner, and an evaluation of the model accuracy and predictive capability.

This work was created in partial fulfillment of Penn State University Capstone Courses “CMPSC 484”. The work is a result of the Psyche Student Collaborations component of NASA’s Psyche Mission ( “Psyche: A Journey to a Metal World” [Contract number NNM16AA09C] is part of the NASA Discovery Program mission to solar system targets. Trade names and trademarks of ASU and NASA are used in this work for identification only. Their usage does not constitute an official endorsement, either expressed or implied, by Arizona State University or National Aeronautics and Space Administration. The content is solely the responsibility of the authors and does not necessarily represent the official views of ASU or NASA.