天文时间序列的分析研究
发布时间：2018-08-07 07:22
【摘要】：随着国内外大型天文观测设备的陆续投入使用,获取的天体数据在时间、空间与频率等方面都得到了极大扩展.可以讲,这是一个天文观测数据爆发的时代.这些数据包含着丰富的信息,如何从海量数据中获取有用的信息,这不光是天文学所面对的挑战,也是其它自然科学与社会科学需要共同应对的难题.我们的研究,聚焦于宇宙中最为极端的变化现象,即Blazar天体的各波段光变,其变化时标分布在从分钟到上百年,变化率也横跨多个数量级,表现出极为丰富的变化模式.我们注意到,在金融、水文、地理、气象、生物、社会等数据分析等领域，也存在着类似的变化现象.并发展出了大量有效的变化模型.我们希望将这些分析方法和模型应用于Blazar天体的光变研究,探寻天体光变中的周期与非周期现象,建立适当的数学模型.并寻求变化的物理原因,对Blazar天体光变的机制做观测的限定,获得合理的物理解释. 我们选取了4个典型的Blazar天体,获得了Fermi天文卫星观测的伽马波段每日观测数据,结合当代时间序列统计方法进行了分析.在光变强度分布方面,我们确认其符合对数正态分布.在变化的内在联系上,其表现为滑动平均时间序列.基于以上分析,我们得到：独立变化存在持续时间,但其影响不会延续太久,数据所表现的长期相关性应该是一种统计意义上的相关.光变强度的对数高斯分布则暗示其变化有可能来自于相对论性喷流中可能的湍流现象. 基于以上尝试,我们认为,面对不断增长的观测数据,使用新型的纯数据分析方法可以得到有数学意义的结果,可以对天体光变模式进行识别和分类.但我们的目的是理解这些变化背后所蕴含的物理.需要理解这些方法的物理意义,并最终与物理模型相结合.
[Abstract]:With the application of large astronomical observation equipments at home and abroad, the data of celestial bodies have been greatly expanded in time, space and frequency. It can be said that this is an era of astronomical observations. These data contain abundant information. How to obtain useful information from massive data is not only a challenge faced by astronomy, but also a difficult problem that other natural sciences and social sciences need to deal with together. Our study focuses on the most extreme phenomena of change in the universe, namely, the light variations in various bands of Blazar celestial bodies, whose time scales range from minutes to hundreds of years, and the rate of change also spans many orders of magnitude, showing extremely rich patterns of variation. We note that there are similar changes in areas such as finance, hydrology, geography, meteorology, biology, and social data analysis. And developed a large number of effective change models. We hope that these analytical methods and models can be applied to the study of the light variation of Blazar objects, to explore the periodic and aperiodic phenomena in the light variation of celestial bodies, and to establish appropriate mathematical models. The physical reasons of the change are sought, and the observation mechanism of the Blazar celestial body is limited, and a reasonable physical explanation is obtained. Four typical Blazar objects are selected and the daily observation data of gamma-band observed by Fermi astronomical satellite are obtained and analyzed with the statistical method of contemporary time series. In terms of light intensity distribution, we confirm that it conforms to the logarithmic normal distribution. In the internal relation of the change, it appears as the moving average time series. Based on the above analysis, we conclude that independent changes exist for a long time, but their effects do not last for long. The long-term correlation of the data should be a statistical correlation. The logarithmic Gao Si distribution of light varying intensity suggests that the variation may be due to the possible turbulence in relativistic jets. Based on the above experiments, we think that in the face of the increasing observation data, the new pure data analysis method can get the mathematical results, and can recognize and classify the celestial light variable patterns. But our aim is to understand the physics behind these changes. The physical implications of these methods need to be understood and eventually combined with physical models.
【学位授予单位】：广州大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：P11;P15
本文编号：2169287
[Abstract]:With the application of large astronomical observation equipments at home and abroad, the data of celestial bodies have been greatly expanded in time, space and frequency. It can be said that this is an era of astronomical observations. These data contain abundant information. How to obtain useful information from massive data is not only a challenge faced by astronomy, but also a difficult problem that other natural sciences and social sciences need to deal with together. Our study focuses on the most extreme phenomena of change in the universe, namely, the light variations in various bands of Blazar celestial bodies, whose time scales range from minutes to hundreds of years, and the rate of change also spans many orders of magnitude, showing extremely rich patterns of variation. We note that there are similar changes in areas such as finance, hydrology, geography, meteorology, biology, and social data analysis. And developed a large number of effective change models. We hope that these analytical methods and models can be applied to the study of the light variation of Blazar objects, to explore the periodic and aperiodic phenomena in the light variation of celestial bodies, and to establish appropriate mathematical models. The physical reasons of the change are sought, and the observation mechanism of the Blazar celestial body is limited, and a reasonable physical explanation is obtained. Four typical Blazar objects are selected and the daily observation data of gamma-band observed by Fermi astronomical satellite are obtained and analyzed with the statistical method of contemporary time series. In terms of light intensity distribution, we confirm that it conforms to the logarithmic normal distribution. In the internal relation of the change, it appears as the moving average time series. Based on the above analysis, we conclude that independent changes exist for a long time, but their effects do not last for long. The long-term correlation of the data should be a statistical correlation. The logarithmic Gao Si distribution of light varying intensity suggests that the variation may be due to the possible turbulence in relativistic jets. Based on the above experiments, we think that in the face of the increasing observation data, the new pure data analysis method can get the mathematical results, and can recognize and classify the celestial light variable patterns. But our aim is to understand the physics behind these changes. The physical implications of these methods need to be understood and eventually combined with physical models.
【学位授予单位】：广州大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：P11;P15
【参考文献】
相关期刊论文 前1条
1 卞维豪,黄克谅;类星体样本光变研究进展[J];天文学进展;1999年03期
，本文编号：2169287
本文链接：https://www.wllwen.com/kejilunwen/tianwen/2169287.html
