基于网格聚类分析LAMOST恒星光谱与理论模型的差异

发布时间：2018-04-24 22:35

本文选题：聚类分析 + Kurucz模型光谱　；参考：《光谱学与光谱分析》2017年06期

【摘要】：随着天文大数据时代计算科学的蓬勃发展,我国具备自主知识产权的国际天文界口径最大、光谱获取率最高的大视场望远镜LAMOST,已率先在国际上开拓并实现了同时观测几千个天体光谱的大规模巡天工作。自2011年巡天至2015年6月所获得的DR3光谱数据集目前已获取世界上最大的恒星参数星表。针对LAMOST第三期发布FGK恒星光谱的流量定标等相关数据,利用Kurucz模板光谱对应的参数空间划分网格,基于开源高效的数据处理R语言程序软件平台,设计了有监督的聚类中心,便于验证其理论参数网格的差异。处理LAMOST实测光谱经归一化后,选择距离量直接描述属性,采用欧氏距离分析判别光谱之间的相似度,选取相应的属性向量构造函数判断观测光谱和理论光谱差别的量级。实验表明:比对LAMOST实测FGK型恒星光谱数据与Kurucz理论模板库数据一致性以及参数测量的准确性,结果显示相同参数的光谱间特征谱线具有较好的一致性,从而得出LAMOST光谱测量物理参数质量较高,具备极好的可靠性,为后续恒星大气模型的改进提供相应的论证依据。
[Abstract]:With the vigorous development of computational science in the era of astronomical big data, China has the largest caliber in the international astronomical circles with independent intellectual property rights. LAMOST, a large field telescope with the highest spectral acquisition rate, has pioneered and realized a large-scale sky survey to simultaneously observe the spectra of thousands of celestial bodies in the world. The DR3 spectral data set from the 2011 survey to June 2015 has now acquired the world's largest star parameter catalog. Aiming at the related data of FGK star spectrum calibration published in the third issue of LAMOST, the supervised clustering center was designed based on the open source and efficient data processing R language software platform by using the parameter space corresponding to the Kurucz template spectrum. It is convenient to verify the difference of the theoretical parameter mesh. After normalization of the LAMOST measured spectrum, the distance quantity is chosen to describe the attribute directly, Euclidean distance analysis is used to distinguish the similarity between the spectra, and the corresponding attribute vector construction function is selected to judge the magnitude of the difference between the observed spectrum and the theoretical spectrum. The experimental results show that the spectral data of FGK type star measured by LAMOST are consistent with the data of Kurucz theoretical template library and the accuracy of parameter measurement. The results show that the spectral characteristic lines of the same parameters have good consistency. It is concluded that the physical parameters measured by LAMOST spectrum are of high quality and excellent reliability, which provides a basis for the improvement of the subsequent stellar atmospheric model.
【作者单位】：哈尔滨工程大学理学院纤维集成光学教育部重点实验室;齐齐哈尔大学;中国科学院光学天文重点实验室(国家天文台);
【基金】：国家自然科学基金项目(U1631239) 黑龙江省自然科学基金项目(F2015024) 黑龙江省教育厅科研项目(135109219)资助
【分类号】：TP311.13

【参考文献】