CNC Lathe With Linear Guideway Slant Bed Desktop Cnc Lathe,Cnc Turning Machine,Cnc Lathe Machine,Slant Bed Cnc Lathe Ningbo junfa CNC Equipment Co. Ltd. , https://www.nbintimecnc.com
[ Instrument network instrument research and development ] Recently, Long Qian, a researcher at the Lijiang Astronomical Observatory of the Yunnan Observatory of the Chinese Academy of Sciences, cooperated with the team of Professor Er Xinzhong from the Cosmology Research Group of the Southwest Institute of Astronomy, Yunnan University, using artificial intelligence deep learning methods to discover 38 new The strong gravitational lens candidate.
The galaxy-scale strong gravitational lens system is an important cosmological probe, which can be used for in-depth study of many scientific issues in cosmology and astrophysics, such as the properties of dark matter, the formation and evolution of galaxies, and the measurement of the Hubble constant. However, the number of certified strong lens systems is too small, which restricts the development of related astrophysics problems.
How to search and verify more strong lens samples is the main problem in current work. Through the development of the next-generation large-scale photometric survey project, people expect to discover tens of thousands of strong lens systems. But how to quickly find strong lens candidates in massive images of celestial bodies? In recent years, the rapid development of artificial intelligence has provided us with new possibilities. Related research teams in the world have used convolutional neural network methods to search for strong gravitational lens systems.
Long Qian has been engaged in the research of artificial intelligence deep learning for a long time, and cooperated with Erxinzhong team to build and train a convolutional neural network. The neural network uses Julia language to be customized according to the characteristics of gravitational lens data. It has small scale, fast speed, Highly targeted characteristics. Researchers applied it to the European Southern Observatory's 2.6-meter Sky Survey Telescope (VST) Kilo-Degree Survey—KiDS data and found 38 new strong lens candidates.
In addition, by testing the performance of the convolutional neural network on different observation conditions and training the network with different sizes of training sets, the researchers tested the stability of the convolutional neural network. The neural network constructed in this research can also be applied to other sky survey data.
Long Qian is the co-corresponding author of the paper. The research work is funded by the General Program of the National Natural Science Foundation of China and the Yunnan Provincial Overseas High-level Talents Program.