SSS巡天星表系统设计和光纤分配算法优化

发布时间：2018-05-02 23:41

本文选题：LAMOST + 大样本巡天　；参考：《中国科学技术大学》2011年博士论文

【摘要】：大天区视场多目标光纤光谱天文望远镜(LAMOST)是国家九五大科学工程之一,是上个世纪由我国以王绶t/、苏定强为首的天文学家根据天文学界大样本光谱观测的需要,创新性的提出的大口径兼备大视场的大型光学望远镜。LAMOST项目1997年由国家计划批准立项,2001年8月正式开工,2008年8月落成,2009年6月4日在中国科学院国家天文台兴隆观测基地顺利通过国家竣工验收,2010年4月17日由国家天文台在兴隆观测站举行了冠名仪式,将LAMOST命名为“郭守敬望远镜”。目前,望远镜即将由试观测阶段进入正式巡天阶段。LAMOST成功突破了大口径与大视场难以兼备的瓶颈,成为国际上继SDSS数字巡天和2dF巡天之后最受瞩目的巡天望远镜,使我国跻身于国际天文技术前沿,成为国际上少数具备自主研制巨型望远镜能力的国家之一。 LAMOST的科学使命是进行大样本光谱巡天。当前大样本天文学已成为天文学发展的一个重要方向。天文学越来越清楚的表明,对于天体物理和宇宙学中的一些重要的问题,例如宇宙的结构、星系的形成和演化,银河系的形成和演化等重要问题都涉及到复杂的物理过程和多方面的影响,这些课题的研究往往需要大样本的统计性质。二十世纪末的技术进步,使得大样本巡天观测成为可能,天文学家能够在合理的时间范围内观测上百甚至上千的天体光谱。大样本天体观测是多波段的,从光学、射电、红外、紫外、X射线以及gamma射线波段都能得到天体各方面的物理信息,而其中光学波段的大样本观测是信息量最大、积累最对、观测技术最成熟的。然而浩瀚宇宙的天体总量难以估计,目前发现的天体已达到百亿量级,在这些天体中已获得光谱的部分只占约万分之一,展开大规模的大样本巡天观测已成为当前的首要工作。大样本天文学越来越受到人们的重视。 LAMOST巡天战略的三个核心课题是星系和类星体的红移巡天、恒星与银河系的观测计划和多波段巡天样本的光学证认,预计获取超过千万颗天体的光谱,研究课题将覆盖宇宙学、星系、类星体、银河系等各个方面。望远镜每次观测量约为4000,一个观测夜可观测数万个天体的光谱,如此大量的观测需要一个自动化的系统安排和部署观测流程。如何协调管理各种类型的科学目标的观测需求,处理各方面主客观约束条件,对巡天区域进行完备覆盖,并选择合适的观测天区,分配目标到光纤,生成观测计划,同时保证整体巡天的尽可能高的观测效率,节约宝贵的天文观测时间,以尽早获得尽可能多的符合科学需求的科学数据产出,是LAMOST的巡天战略系统(SSS)的核心科学目标。为了实现能服务于LAMOST巡天模式的软件系统,本论文首先在小系统模型的基础上,分析了大系统的相关环境,采用领域驱动设计的软件设计思想,结合多种软件设计模式,改进了巡天战略系统的静态系统结构,调整了动态工作流程,使得软件满足大系统各方面的需求。巡天战略系统的关键问题是天区覆盖和光纤分配算法。本论文在原有的天区覆盖算法基础上,解决了算法在高纬度处的一些难题,提高算法速度和覆盖效果,应用最大密度思路实现亮星中心覆盖算法;结合望远镜光纤单元具体构造,改进光纤分配算法,避免光纤单元之间的碰撞,建立动态分组以实现光纤配额,支持目标重分配以提高光纤分配效率,体现用户多样化的优先策略。巡天战略系统的核心是观测目标。本论文通过分析LAMOST的具体科学目标,从天文研究人员的选源、观测策略的指定、星表管理、观测计划生产过程中的目标过滤到观测结果反馈,为SSS建立一套全面的面向分布式环境的软件环境。为了进一步验证软件的具体功能和望远镜的巡天能力,本论文根据一组实际观测时间数据,引入SDSS巡天的观测星表,进行了基于实际观测时间的大天区巡天观测模拟,以验证巡天战略系统算法的有效性,并对望远镜的巡天观测能力进行评估。巡天战略系统是连接天文学家和望远镜观测的桥梁,本论文结合各方用户的需求,改进用户层设计,增加功能覆盖范围,扩大软件的应用领域,以期将系统扩展为一款天文通用软件。最后,本论文还将就SSS的下一步的发展方向进行讨论。本论文的创新点在于: 1、采用领域驱动设计的软件开发方法对巡天战略系统进行升级,提高了系统的扩展性和健壮性,改进了用户接口,建立了一个与LAMOST望远镜观测相结合的模拟星空平台以呈现天体观测位置和周日运行轨迹,完成了一套完整的基于LAMOST望远镜的大样本巡天战略系统,已用于LAMOST的观测中。 2、结合望远镜实际特点和需求,优化了天区覆盖算法模型,建立密度库算法,引入快速漂移算法,设计边界漂移算法和SH亮星覆盖算法;根据焦面光纤单元的具体构造,建立光纤分配算法,避免光纤单元碰撞,保证天光和定标星在焦面上的分配需求,支持分组和目标重分配。 3、对LAMOST望远镜巡天观测选源的需求进行分析,以观测星表数据库为基础,建立了一个面向网络环境的通用数据库类型访问、星表数据选择、数据格式转换和星表录入平台;在此基础上设计和实现目标过滤模型,根据各种观测约束条件选择观测目标;设计了和实现了观测计划数据库及其支持组件。
[Abstract]:LAMOST is one of the nine five major scientific projects in the country . It is one of the nine five major scientific projects in China . It is a large - scale optical telescope with large aperture and large field of view in the past century . The LAMOST project was approved by the State Plan in 1997 . The LAMOST project was officially launched in August , 2008 .

The scientific mission of LAMOST is to conduct a large sample spectrum tour . The current large sample astronomy has become an important direction in the development of astronomy .

The three core topics of the LAMOST tour strategy are the red shift of galaxies and quasars , the observation plan of stars and the Milky Way , and the optical identification of multi - band patrol samples . The telescope is expected to acquire the spectra of more than 10 million celestial bodies . Such a large number of observations require an automated system arrangement and deployment observation process . How to coordinate the management of the observation needs of various types of scientific objectives , allocate the target to the optical fiber , generate observation plans , and save valuable astronomical observation time , so as to obtain as much scientific data output as possible , which meets the scientific needs as much as possible , is the core scientific objective of the tour strategic system ( SSS ) of LAMOST .

In order to realize the software system capable of serving the LAMOST patrol mode , this paper first analyzes the related environment of the large system on the basis of the small system model , adopts the software design idea of the field driving design , and combines various software design modes , improves the static system structure of the patrol strategic system , and adjusts the dynamic workflow , so that the software can meet the requirements of all aspects of the large system .

On the basis of the original method , the paper solves some problems of the algorithm in the high latitude , improves the speed and the coverage effect of the algorithm , and applies the maximum density to realize the bright star center covering algorithm , and combines the specific structure of the telescope optical fiber unit , improves the fiber distribution algorithm , avoids collision between the fiber units , establishes dynamic grouping to realize the fiber quota , and supports the target redistribution to improve the distribution efficiency of the optical fiber and reflects the priority strategy of the user diversification .

Through analyzing the specific scientific objectives of LAMOST , this paper analyzes the specific scientific objectives of LAMOST , from the source of the astronomical researchers , the designation of the observation strategy , the star table management , the objective filtering in the production process of the observation plan to the observation result feedback , and establishes a comprehensive software environment for the SSS .

In order to further verify the specific function of the software and the inspection ability of the telescope , this paper introduces the observation satellite table of SDSS patrol according to a set of actual observation time data , and carries out the simulation of the daily observation of the large - day area based on the actual observation time to verify the effectiveness of the algorithm of the patrol - day strategic system and evaluate the inspection ability of the telescope .

The system is a bridge connecting astronomers and telescopes . This thesis combines the needs of all parties , improves user - level design , increases functional coverage , and expands the application fields of software , with a view to extending the system into an astronomical general software .

Finally , this thesis will discuss the next step of SSS .

The innovation point of this thesis lies in :

1 . The software development method in the field driving design is adopted to upgrade the patrol strategic system , the expansibility and robustness of the system are improved , the user interface is improved , a simulated star - space platform combined with the LAMOST telescope observation is established to present the celestial observation position and the Sunday running track , and a complete set of large sample patrol strategic system based on the LAMOST telescope is completed , and has been used in the observation of LAMOST .

2 , combining the actual characteristics and requirements of the telescope , optimizing the sky coverage algorithm model , establishing a density library algorithm , introducing a fast drift algorithm , designing a boundary drift algorithm and an SH bright star coverage algorithm , establishing an optical fiber distribution algorithm according to the specific configuration of the focal plane optical fiber unit , avoiding collision of the optical fiber unit , ensuring the distribution requirements of the sky light and the calibration star on the focal plane , supporting the grouping and the target redistribution .

3 . Based on the observation satellite table database , a universal database type access , star table data selection , data format conversion and star table entry platform for the network environment are established based on the observation satellite table database . Based on this , the target filtering model is designed and realized , and the observation target is selected according to various observation constraints ; and the observation plan database and its support components are designed and realized .

【学位授予单位】：中国科学技术大学
【学位级别】：博士
【学位授予年份】：2011
【分类号】：TH751

【参考文献】