Space-use Hall electric thruster developed with AI [Now Science]
The Hall thruster is a high-efficiency propulsion device using plasma that is used in various difficult space missions such as SpaceX's Starlink satellite constellation and NASA's Psyche asteroid probe. It is one of the core space technologies.
A research team at the Korea Advanced Institute of Science and Technology (KAIST) has developed a CubeSat Hall thruster using AI techniques. It will be installed on the CubeSat K-HERO in the fourth launch of Nuriho scheduled for November of this year to verify its performance in space.
The Hall thruster developed by the research team is scheduled to be installed on the CubeSat (K-HERO) that will be launched into space in the second half of this year through the fourth launch of Nuriho to perform in-orbit verification. [Photo = KAIST]
KAIST (President Lee Kwang-hyung) announced on the 3rd that the research team led by Professor Choi Won-ho of the Department of Nuclear and Quantum Engineering has developed an artificial intelligence technique that can predict the thrust performance of Hall electric thrusters (Hall thrusters), which are engines for satellites and space probes, with high accuracy.
Hall thrusters have high fuel efficiency, so they can greatly accelerate satellites and spacecraft using less propellant (fuel). They can generate large thrust compared to the power consumed. Based on these advantages, they are widely used in various missions such as maintaining formation flights of satellite constellations in space environments where propellant conservation is important, orbital deorbit maneuvers for reducing space debris, and providing propulsion for deep space exploration such as comet or Mars exploration.
As the space industry expands in the NewSpace era, space missions are becoming more diverse. The demand for Hall thrusters is increasing. In order to quickly develop high-efficiency Hall thrusters optimized for each unique mission, a technique that accurately predicts the performance of the thruster from the design stage is essential.
Existing methods have limitations in that they cannot precisely handle the complex plasma phenomena occurring in a Hall thruster or are limited to specific conditions, resulting in low performance prediction accuracy.
The research team developed a highly accurate thruster performance prediction technique based on artificial intelligence that drastically reduces the time and cost required for the repetitive work of designing, manufacturing, and testing a Hall thruster.
Professor Won-ho Choi's team, which was the first to start research on electric thruster development in Korea in 2003 and has been leading related research and development, introduced an artificial neural network ensemble structure based on 18,000 Hall thruster learning data generated using a self-developed electric thruster computational analysis tool and applied it to thrust performance prediction. The
computational analysis tool developed to secure high-quality learning data models plasma physical phenomena and thrust performance. The accuracy of the computational analysis tool was verified to be highly accurate, with an average error of less than 10% compared to approximately 100 experimental data performed with 10 Hall thrusters developed by the research team for the first time in Korea.
The learned artificial neural network ensemble model functions as a digital twin model that can predict the performance of a Hall thruster in a short time, within seconds, with high accuracy depending on the design variables of the thruster.
It can analyze in detail the changes in performance indicators, such as thrust and discharge current, according to design variables, such as fuel flow rate and magnetic field, which were difficult to analyze with previously known scaling laws.
The research team explained that the artificial neural network model developed this time showed an accuracy of within 5% in average error for the 700 W and 1 kW Hall thrusters developed in-house, and within 9% in average error for the 5 kW high-power Hall thruster developed by the U.S. Air Force Research Laboratory.
It was proven that the artificial intelligence prediction technique developed in this study can be widely applied to Hall thrusters of various power sizes.
Professor Choi Won-ho said, "The AI-based performance prediction technique developed by the research team has high accuracy and is already being used in the analysis of the thrust performance of Hall thrusters, which are the engines of artificial satellites and spacecraft, and in the development of high-efficiency, low-power Hall thrusters." He continued, "This AI technique can be applied not only to Hall thrusters, but also to the research and development of ion beam sources used in various industries such as semiconductors, surface treatment, and coating.
" He added, "The CubeSat Hall thruster developed using AI techniques in collaboration with Cosmobee, an electric propulsion specialist and a laboratory startup of the research team, is scheduled to be loaded onto K-HERO, a 3U (30x10x10cm) CubeSat, in the fourth launch of Nuri scheduled for November of this year to verify its performance in space."
The results of this study (paper title: Predicting Performance of Hall Effect Ion Source Using Machine Learning), in which KAIST Department of Nuclear and Quantum Engineering (Space Exploration Engineering Interdisciplinary Major) Ph.D. student Jaehong Park participated as the first author, were published online in the international academic journal 'Advanced Intelligent Systems' on December 25, 2024.
https://www.inews24.com/view/blogger/1809371
댓글
댓글 쓰기