The reporter learned from the Chinese Academy of Sciences that recently, an international team led by Professor Ge Jian of the Shanghai Observatory of the Chinese Academy of Sciences used artificial intelligence to successfully discover 5 stars with diameters smaller than the Earth and orbits in the photometric data of stars released by the Kepler Space Telescope in 2017. Ultra-short-period planets with periods shorter than 1 day, four of which are the smallest planets discovered so far that are closest to their host stars, are similar in size to Mars. This is the first time that astronomers have used artificial intelligence to complete the task of searching for suspected signals and identifying real signals at once. The relevant research results were published in the recently published international astronomy journal Monthly Notices of the Royal Astronomical Society (MNRAS).
Ultra-short-period exoplanets were first discovered in the Kepler space telescope photometry data in 2011, bringing unique opportunities and challenges to planet formation theory, prompting scientists to re-examine and improve existing planetary system formation and evolution models. .
Ge Jian said that the existence of ultra-short-period planets provides important clues for the study of the early evolution of planetary systems, planet-planet interactions, and the dynamics of star-planet interactions (including tidal forces and atmospheric erosion). For example, ultrashort-period planets likely did not form at their current locations but migrated inward from their original orbits—this is because the host stars of these ultrashort-period planets had much larger radii and farther distances during their early formation stages than they do today. If the nearest ultra-short-period planet is near the star during the star formation stage, it is likely that it will have been swallowed by its host star. "Also, given that ultrashort-period planets are often observed accompanied by outer planets in longer-period orbits, it is speculated that the origin of ultrashort-period planets involves interactions between planetary siblings that relocate the ultrashort-period planets to They are currently in orbits close to their host star, which may have been orbits previously occupied by the star itself. Alternatively, such orbital migration of ultrashort-period planets may be caused by interactions with the protoplanetary disk or through tides with the host star. Driven by interaction.”
So far, humans have only found a total of 145 ultra-short period planets, of which only 30 have radii smaller than the radius of the Earth. Ge Jian said: "We still know very little about ultra-short period planets, because the sample size is too small, and it is difficult to accurately understand their statistical characteristics and occurrence rates."
The new research provides a new way to search for ultra-short-period planets. Ge Jian said: "Inspired by Professor Li Xiaolin from the Department of Computer Science at the University of Florida, I tried to apply deep learning of artificial intelligence to the photometry data released by the Kepler space telescope to find faint transiting stars that could not be found using traditional methods. Signal. After nearly 10 years of hard work, we finally have it. The first harvest. If you want to use artificial intelligence to 'dig' extremely rare new discoveries in massive astronomical data, you need to develop innovative artificial intelligence algorithms and conduct specific training so that they can explore quickly, accurately, and completely. to these rare and weak signals that are difficult to find using traditional methods.”
Princeton University astrophysicist Josie Wing commented: "Ultra-short-period planets, or 'lava worlds,' have unexpected properties that provide clues to our understanding of how planetary orbits change over time. This technology to find new planets I am impressed by the achievement.”