本文選題：郭守敬望遠鏡(LAMOST) + 天體光譜　； 參考：《山東大學》2011年碩士論文

【摘要】：天文學是一門古老的科學,自有人類文明史以來,天文學就有重要的地位。觀測儀器設(shè)備及數(shù)據(jù)收集能力的大幅度提高,使得我們邁入了天文觀測數(shù)據(jù)的“雪崩”時代。天體在光學波段的光譜包含著豐富的物理信息。星系的光譜可以給出它們的距離、構(gòu)成、分布和運動等信息。恒星的光譜可以確定它們的分布和運動、光度、溫度、化學組成等物理狀態(tài)。從大量天體的光譜觀測中還會發(fā)現(xiàn)奇異的天體和天文現(xiàn)象,將引起人類對宇宙天體的新認識。 LAMOST巡天正式開始,每晚的觀測將要產(chǎn)生數(shù)萬條光譜。整個巡天計劃完成將會產(chǎn)生107數(shù)量級的光譜數(shù)據(jù),如此龐大數(shù)量的光譜顯然不能通過傳統(tǒng)的人工方式進行的,因此需要研究相關(guān)的算法進行光譜的自動處理。 在一維光譜處理中,恒星參數(shù)測量部分為銀河系恒星光譜巡天提供恒星運動學、化學豐度、有效溫度、有效重力加速度等信息,是一維光譜處理中非常重要的部分。從海量巡天數(shù)據(jù)發(fā)現(xiàn)特殊稀少未知天體,能夠為天文學各種研究提供樣本支持。 本研究的工作主要分為兩部分： (1)設(shè)計并實現(xiàn)適用于LAMOST光譜的恒星大氣參數(shù)測量系統(tǒng),包括光譜預(yù)處理、參數(shù)測量等模塊,其中參數(shù)測量模塊支持擴展性,現(xiàn)已集成SSPP、UlySS等軟件包,并加入PLS方法,隨著研究的不斷進行,會有更多的方法集成進來。軟件系統(tǒng)采用Python,結(jié)合GTK實現(xiàn)圖形用戶界面,運用多線程編程計算實現(xiàn)對海量光譜的快速批量處理。 (2)研究可用于LAMOST光譜中發(fā)現(xiàn)特殊天體的數(shù)據(jù)挖掘的算法,包括有指導(dǎo)和無指導(dǎo)兩類,其中前者主要發(fā)現(xiàn)一些已知的特殊天體,而后者主要是發(fā)現(xiàn)一些未知的特殊天體。研究的方法主要包括隨機森林算法及遺傳算法等。
[Abstract]:Astronomy is an ancient science, since the history of human civilization, astronomy has played an important role. With the great improvement of observational equipment and data collection ability, we have entered the "avalanche" era of astronomical observation data. The spectra of celestial bodies in the optical band contain abundant physical information. The spectra of galaxies can give information about their distance, composition, distribution, and motion. The spectra of stars determine their distribution and motion, luminosity, temperature, chemical composition and other physical states. Strange celestial bodies and astronomical phenomena will also be found from the spectral observations of a large number of celestial bodies, which will lead to a new understanding of the celestial bodies in the universe. The LAMOST survey officially begins, and observations each night will produce tens of thousands of spectra. The completion of the whole survey plan will produce 107 order of magnitude spectral data. It is obvious that such a large number of spectra can not be carried out by traditional manual methods, so it is necessary to study related algorithms for spectrum automatic processing. In the one-dimensional spectral processing, the measurement of star parameters provides the information of star kinematics, chemical abundance, effective temperature and effective gravity acceleration for the spectral survey of galactic stars, which is a very important part of one-dimensional spectral processing. Special rare unknown bodies can be found from massive survey data, which can provide sample support for various astronomical studies. The work of this study is divided into two parts: 1) designing and implementing the stellar atmospheric parameter measurement system suitable for LAMOST spectrum, including spectral preprocessing, parameter measurement and other modules, in which the parameter measurement module supports expansibility. The software package SSPPU UlySS has been integrated, and the PLS method has been added to the system. As research continues, more methods will be integrated. The software system uses Python and GTK to realize the graphical user interface. The multithread programming is used to realize the fast batch processing of the mass spectrum. This paper studies the data mining algorithms for discovering special objects in the LAMOST spectrum, including guided and unguided data mining, in which the former mainly finds some known special objects, while the latter mainly finds some unknown special celestial bodies. The research methods mainly include stochastic forest algorithm and genetic algorithm.
【學位授予單位】：山東大學
【學位級別】：碩士
【學位授予年份】：2011
【分類號】：TP274;P111

【引證文獻】

相關(guān)碩士學位論文 前2條

1 劉杰;基于模板匹配的恒星大氣物理參數(shù)自動測量的研究[D];山東大學;2012年

2 林雪梅;ANN在天體光譜分類及恒星大氣參數(shù)測量中的應(yīng)用[D];山東大學;2012年

，

本文編號：1779855