KAIST researchers will validate the performance of the Hall thruster, developed using artificial intelligence (AI) techniques for cube satellites, during the Nuri rocket's fourth launch scheduled for November this year.
KAIST announced on the 3rd that Professor Choi Won-ho's research team of the Department of Nuclear and Quantum Engineering developed an AI technique that can accurately predict the thrust performance of Hall thrusters, which are the engines for satellites or space exploration vehicles.
Hall thrusters are high-efficiency propulsion devices that use plasma, allowing for significant acceleration of satellites or spacecraft while consuming little fuel. They are widely used for various missions, including maintaining the formation flying of SpaceX's Starlink satellites and providing thrust for comet or Mars exploration.
To develop a high-efficiency Hall thruster, it is essential to accurately predict the performance of the thruster from the design stage. However, existing methods have limitations, as they do not precisely handle the complex plasma phenomena occurring within the Hall thruster or have low accuracy in performance prediction limited to specific conditions.
To address this, the research team developed its own AI-based technique for predicting thruster performance. This significantly reduced the time and expense required for the repetitive processes of designing, manufacturing, and testing Hall thrusters. Additionally, they created 18,000 learning data points using their self-developed computational analysis tools, applying them to the AI model for more accurate performance predictions.
The actual performance showed minimal error. When comparing the actual performance with experimental data for 10 domestically developed Hall thrusters, the average error was within 10%. The research team demonstrated an average error of within 5% for the 700W and 1kW Hall thrusters they developed, and an average error of within 9% for the 5kW high-power Hall thruster developed by the U.S. Air Force Research Laboratory. This research proved that the AI prediction technique developed can be broadly applied to various Hall thrusters.
Professor Choi Won-ho noted, "This artificial intelligence technique can be applied not only to Hall thrusters but also to the research and development of ion beam sources used in various industries, including semiconductors, surface treatment, and coatings."