Shell stars, in particular the cooler ones, often do not show conspicuous Balmer-line emission and may consequently be missed in surveys that specifically search for emission signatures in the Halpha line. The present work is aimed at identifying stars with shell-signatures via a search for strong FeII multiplet 42 lines at 4924, 5018, 5169A in archival LAMOST spectra. Candidates were selected by probing the FeII 42 lines in the spectra of a sample of colour-preselected early-type stars using a modified version of the MKCLASS code and then categorised by visual inspection of their spectra. We identified 75 stars showing conspicuous shell features, 43 Am/CP1 stars, 12 Ap/CP2 stars, and three objects with composite spectra. Spectral types and equivalent width measurements of the FeII 42 lines are presented for the sample of shell stars. Except for three objects, all shell stars appear significantly removed from the ZAMS in the colour-magnitude diagram, which is likely due to extinction by circumstellar material. We find a correlation between the equivalent width of the 5169A line and the distance to the locus of the main-sequence stars (the larger the IR-excess, the stronger the 5169A line) and studied the variability of the shell star sample using TESS data, identifying a very high proportion of double stars. All but 14 shell stars are new discoveries, which highlights the efficiency of the here presented novel approach to identify stars with subtle shell features. This study may be used as a blueprint for discovering these objects in massive spectral databases.
Searching for Contact Binaries with LAMOST and TESS
Xinjiang Astronomical Observatory, Chinese Academy of Sciences（中国科学院新疆天文台）；School of Astronomy and Space Science, University of Chinese Academy of Sciences（中国科学院大学天文学与空间科学学院）；Department of Astronomy, Xiamen University（厦门大学天文学系）；College of Physics and Information Engineering, Quanzhou Normal University（泉州师范学院物理与信息工程学院）
The Spectroscopic Binaries from LAMOST Medium-Resolution Survey (MRS). I. Searching for Double-lined Spectroscopic Binaries (SB2s) with Convolutional Neural Network
Metallic-Line Stars Identified from Low Resolution Spectra of LAMOST DR5
LAMOST DR 5发布了超过20万个S/N>50的早期型恒星的低分辨率光谱。本文介绍了在这样一个大型数据库中寻找金属线星并研究其统计性质的工作。使用已知的Am光谱试验了六种机器学习算法，还研究了经验标准方法（Hou等人，2015）和MKCLASS包（Gray等人，2016）。比较它们的性能，随机森林（RF）算法获胜，不仅因为RF具有高的成功率，而且它可以推导和排名的功能。然后对DR 5的早型星进行了射频测量，从中筛选出15,269颗Am候选星。根据RF算法得出的光谱特征进行人工识别，并由专家进行验证。经过人工识别，9，372颗Am星和1，131颗Ap候选星被编入星表。进行了统计研究，包括温度分布，空间分布，和红外测光。Am星的光谱类型主要在F0-A4之间，在A7附近有一个峰值，这与以前的工作相似。根据盖亚距离，我们计算了每颗Am恒星从银河平面到银河平面的垂直高度Z。Z的分布表明Am星的入射率随jZj的增大呈下降梯度。另一方面，Am星在红外波段没有显示出值得注意的模式。随着波长变长，Am星的红外过量减少，直到W1和W2波段很少或没有过量。
Minute-cadence Observations of the LAMOST Fields with the TMTS: I. Methodology of Detecting Short-period Variables and Results from the first-year Survey
