以機器學習法同時搜尋系外行星與變星之研究

簡易檢索 / 詳目顯示

回結果列表

研究生：	程麒融 Cheng, Chi-Jung
論文名稱：	以機器學習法同時搜尋系外行星與變星之研究 Searching Exoplanets and Variable Stars Simultaneously through Machine Learning
指導教授：	葉麗琴 Yeh, Li-Chin
口試委員:	江瑛貴 Jiang, Ing-Guey 陳賢修 Chen, Shyan-Shiou
學位類別：	碩士 Master
系所名稱：	理學院 - 計算與建模科學研究所 Institute of Computational and Modeling Science
論文出版年：	2023
畢業學年度：	111
語文別：	英文
論文頁數：	70
中文關鍵詞：	機器學習、系外行星、變星
外文關鍵詞：	CoRoT, VariablesStar
相關次數：	點閱：3 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

系外行星以及變星的分類是太空科學當中相當重要的一份工作。在此文章中，我們試著用機器學習的方法去分類系外行星以及變星。我們在此文章中所使用的是CoRoT太空望遠鏡資料庫。這些資料在未進行處理之前是很難讓機器學習去做辨認的，所以資料處理也是此文章中相當重要的一份工作。
模型的穩定度以及機器學習的結果將呈現在文章的後半部。下一步我們用Mandel & Agol (2002）[5]和Bhardwaj.A et al (2016）[1]構築理論光曲線以計算其與凌星光曲線的誤差以及和變星光曲線的誤差。最後我們在此資料庫中成功找到5個後變星以及8個候選行星。

Classification of exoplanets and variable stars has been a vital work in space science. In this work, we try to classify exoplanets and variable stars simultaneously by using machine learning methods. The data we focused in this work is from CoRoT space mission. The data from CoRoT isn’t easy for the machine learning to classify, so data analyzing is also important in our work. The stability and the results of our models are presented as well. Next, we will construct the theoretical light curves by Mandel&Agol(2002）[5] and Bhardwaj.A et al(2016）[2] in order to calculate the errors between theoretical light curves and observational light curves. Finally, we found five possible variable stars and eight possible exoplanets from our CoRoT dataset.

Abstract............................................................i
Acknowledgement....................................................iii
CHAPTER 1 Introduction..............................................1
CHAPTER 2 Data Analysis.............................................3
2.1 CoRoT Data Sets...............................................3
2.2 Generating a Noise Light Curve................................5
2.2.1 Normalization.............................................6
2.2.2 Linear Regression.........................................6
2.2.3 Folding...................................................9
2.2.4 Interpolation............................................10
CHAPTER 3 Models and CNN Structure.................................12
3.1 Generating Transit Signal Light Curves.......................12
3.2 Variable Star Model..........................................17
3.3 CNN Structure................................................21
3.3.1 Our CNN model............................................21
3.3.2 Optimizers and Loss Function.............................25
3.3.3 Early Stopping...........................................25
3.3.4 K-fold Cross Validation..................................25
CHAPTER 4 Results..................................................27
4.1 Constructing Models for Different Type of Stars..............27
4.2 Generating Sata Sets for Machine Learning....................30
4.3 Choosing the Proper Amount of Data Sets for Each Model.......30
4.4 Evaluation Matrices..........................................38
CHAPTER 5 Possible Exoplanets and Variable Stars...................52
5.1 Possible Exoplanets Candidates...............................53
5.2 Possible Variable-Star Candidates............................58
CHAPTER 6 Conclusion...............................................61
Reference..........................................................62
Appendix...........................................................64

                                

[1] Aguirre .C,1K. Pichara & I. Becker1(2019）。 Deep multi-survey classification of variable stars. Monthly Notices of the Royal Astronomical Society, Volume 482, Issue 4, Pages 5078–5092
[2] Bhardwaj.A et al (2016）。 A comparative study of multiwavelength theoretical and observed light curves of Cepheid variables. Monthly Notices of the Royal Astronomical Society, Volume 466, Issue 3, April 2017, Pages 2805–28243
[3] Bhardwaj.A et al (2015）。 On the variation of Fourier parameters for Galactic and LMC Cepheids at optical, near-infrared and mid-infrared wavelengths. Monthly Notices of the Royal Astronomical Society, Volume 447, Issue 4,11March 2015, Pages 3342–3360
[4] Hossin, M.1&Sulaiman, M.N (2015）。 A review on evaluation metrics for data classification evaluations Process 5(2）：01-11
[5] Mandel, K.,&Agol, E(2002）。 Analytic Lightcurves for Planetary Transit Searches. The Astrophysical Journal, Volume 580, Issue 2, pp. L171-L175.
[6] Yeh, L.-C., Jiang, I.-G(2021）。 Searching for Possible Exoplanet Transits from BRITE Data through a Machine Learning Technique. Publications of the Astronomical Society of the Pacific [SCI]，133:014401(12pp)。
[7] https://sci.esa.int/web/corot/-/31709-summary
[8] https://en.wikipedia.org/wiki/Stellar_classification

簡易檢索 / 詳目顯示

相關論文