跳到主要內容

簡易檢索 / 詳目顯示

回結果列表
研究生： 邱思齊
Shih-Chi Chiu
論文名稱： 應用長短期記憶遞迴神經網路預測Kp地磁指數
Forecasting of Kp Index Using Long Short-Term Memory Recurrent Neural Networks
指導教授： 許志浤
Jih-Hong Shue
口試委員:
學位類別： 碩士
Master
系所名稱： 地球科學學院 - 太空科學與工程學系
Department of Space Science and Engineering
論文出版年： 2022
畢業學年度： 110
語文別： 中文
論文頁數： 94
中文關鍵詞： Kp指數 、太陽風電漿 、行星際磁場 、長短期記憶 、磁層頂位置
外文關鍵詞： Kp Index, Solar Wind Plasma, Interplanetary Magnetic Field, Long Short-Term Memory, Magnetopause Position
相關次數： 點閱：12  下載：0 
分享至:
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

    • Kp指數是用於檢測地磁擾動水平的最重要指標之一，地磁擾動受驅動於太陽風與磁層之間的交互作用。不同情況的太陽風與磁層狀態的耦合使得Kp隱含著多種未知的時變模式。過去的深度學習模型幾乎都使用標準前饋式神經網路預測Kp，然而由於磁暴期間(Kp ≥ 5) 的參數變化劇烈，並且資料相對較少，前饋式神經網路預測磁暴期間Kp的準確度會被局限。本研究使用長短期記憶來預測Kp，長短期記憶是遞迴神經網路的一種變形，能夠儲存隱含在不同輸入特徵之中長期與目標的時間序列相依性。在本研究中，我們設定模型訓練時每次通過當前時步的神經元都包含過去二十四小時的輸入特徵資訊，成功將預測一小時後Kp的準確度比過去研究增加六個百分比。為了滿足不同的需求與條件，本研究也發展了不使用歷史Kp作為輸入參數的模型，並且使用磁層頂日下點對峙距離與磁尾張開程度的參數取代Kp作為模型輸入。結果顯示輸入磁層頂位置參數能有效地降低使用歷史Kp作為模型輸入的依賴性，也能稍微提升磁暴期間的Kp預測準確度。由於磁暴期間較大的地磁擾動會影響地球上的通信設備和電網系統，因此提高Kp的預測準確度有助於太空天氣預報和預警系統的發展，且能有效降低突如其來的極端太空天氣對人類生活的影響。

    The Kp index is one of the most important indicators used to monitor the level of the geomagnetic disturbances driven by interactions between the solar wind and the magnetosphere. Different coupling modes of the solar wind-magnetosphere are hidden in time-varying Kp, which makes Kp hard to be forecasted. Most of the previous studies used the general feedforward neural networks (FNN) to predict Kp. However, the accuracy of the FNN is limited to the highly variable and sparse nature in parameters during magnetic storms (Kp ≥ 5). This study uses the long short-term memory (LSTM) to predict the Kp. The LSTM is a variant of the recurrent neural networks (RNN) that can store long-term time series dependencies hidden in different input features. We passed the input feature information for the past twenty-four hours to all neurons during each epoch of model training, successfully increasing the accuracy of predicting Kp for one-hour ahead by 6%. This study also developed a model that did not use historical Kp as an input parameter. We used the parameters of the subsolar distance of the magnetopause and the level of its tail flaring to replace Kp as the model input. Our results show that inputting the magnetopause position parameters can reduce the dependence of using historical Kp as the model input, and the accuracy of a Kp prediction during magnetic storms is slightly increased. Since the large geomagnetic disturbance will affect the communication equipment and power grid systems on the earth, improving the accuracy of the Kp prediction will help develop space weather forecasting and warning systems, which can effectively reduce the impact of extreme space weather on human life.

    摘要....................i Abstract....................ii 誌謝....................iii 目錄....................v 圖目錄....................vii 表目錄....................ix 符號說明....................x 第一章 緒論....................1 1-1 研究背景與動機 - 太空天氣與Kp地磁指數....................1 1-2 研究文獻回顧 - Kp預測模型發展歷程....................8 1-3 研究目的與流程....................12 第二章 數據分析及預處理....................14 2-1 數據取得 – OMNI資料庫....................14 2-1-1 原始數據集 – 模型輸入選擇....................16 2-2 數據預處理....................19 2-2-1 數據填補 – 缺失值處理....................19 2-2-2 數據觀察 – 機率密度分布....................22 2-2-3 離群值定義 – 四分位距與盒形圖....................25 2-2-4 特徵縮放 (Feature Scaling) – 離群值處理....................29 第三章 模型訓練與評估....................35 3-1 人工神經網路 (Artificial Neural Network)....................35 3-1-1 神經網路原理....................35 3-1-2 前饋式神經網路預測Kp的缺陷....................36 3-2 遞迴神經網路 (Recurrent Neural Network)....................37 3-2-1 RNN網路構造與網路層....................38 3-2-2 標準RNN用於預測Kp的局限性....................39 3-3 長短期記憶單元 (Long Short-Term Memory)....................41 3-3-1 遺忘閥 (Forget Gate)....................44 3-3-2 輸入閥 (Input Gate)....................45 3-3-3 LSTM記憶單元狀態更新....................46 3-3-4 輸出閥 (Output Gate)....................47 3-4 訓練、驗證與測試集劃分....................47 3-5 不同LSTM模型架構下的Kp預測能力....................49 第四章 Kp預測結果與分析....................53 4-1 磁層頂公式....................53 4-2 模型輸入包含Kp (一小時模型)....................56 4-2-1 模型I (輸入Kp、Tp、Np、Vsw、Pd、Bx、By與Bz)....................56 4-2-2 模型II (輸入Kp、Tp、Np、Vsw、Bx、By、r_0與α)....................61 4-3 模型輸入不包含Kp (一小時模型)....................63 4-3-1 模型III (輸入Tp、Np、Vsw、Pd、Bx、By與Bz)....................63 4-3-2 模型IV (輸入Tp、Np、Vsw、Bx、By、r_0與α)....................65 4-4 模型V – 三小時模型....................67 第五章 討論與結論....................70 5-1 討論....................70 5-2 結論....................70 參考文獻....................73

    Ayala Solares, J. R., Wei, H.-L., Boynton, R. J., Walker, S. N., & Billings, S. A. (2016). Modeling and prediction of global magnetic disturbance in near-Earth space: A case study for Kp index using NARX models. Space Weather, 14(10), 899-916. https://doi.org/10.1002/2016SW001463

    Baker, D. N., Hones Jr., E. W., Payne, J. B., & Feldman, W. C. (1981). A high time resolution study of interplanetary parameter correlations with AE. Geophysical Research Letters, 8(2), 179-182. https://doi.org/10.1029/GL008i002p00179

    Bala, R., & Reiff, P. (2012). Improvements in short-term forecasting of geomagnetic activity. Space Weather, 10(6). https://doi.org/10.1029/2012SW000779

    Bala, R., Reiff, P. H., & Landivar, J. E. (2009). Real-time prediction of magnetospheric activity using the Boyle Index. Space Weather, 7(4). https://doi.org/10.1029/2008SW000407

    Balikhin, M. A., Boaghe, O. M., Billings, S. A., & Alleyne, H. S. C. K. (2001). Terrestrial magnetosphere as a nonlinear resonator. Geophysical Research Letters, 28(6), 1123-1126. https://doi.org/10.1029/2000GL000112

    Bartels, J. (1949). The standardized index Ks, and the planetary index Kp, IATME Bull., 12 (b), 97, IUGG Publ. Office, Paris.

    Bartels, J. (1957). The technique of scaling indices K and Q of geomagnetic activity. Ann. Intern. Geophys., 4, 215-226.

    Bartels, J., Heck, N. H., & Johnston, H. F. (1939). The three-hour-range index measuring geomagnetic activity. Terrestrial Magnetism and Atmospheric Electricity, 44(4), 411-454. https://doi.org/10.1029/TE044i004p00411

    Bartels, J., Heck, N. H., & Johnston, H. F. (1940). Geomagnetic three-hour-range indices for the years 1938 and 1939. Terrestrial Magnetism and Atmospheric Electricity, 45(3), 309-337. https://doi.org/10.1029/TE045i003p00309

    Bennett, D. A. (2001). How can I deal with missing data in my study? Australian and New Zealand journal of public health, 25(5), 464-469.

    Boaghe, O. M., Balikhin, M. A., Billings, S. A., & Alleyne, H. (2001). Identification of nonlinear processes in the magnetospheric dynamics and forecasting of Dst index. Journal of Geophysical Research: Space Physics, 106(A12), 30047-30066. https://doi.org/10.1029/2000JA900162

    Boberg, F., Wintoft, P., & Lundstedt, H. (2000). Real time Kp predictions from solar wind data using neural networks. Physics and Chemistry of the Earth, Part C: Solar, Terrestrial & Planetary Science, 25(4), 275-280.

    Borovsky, J. E., Thomsen, M. F., Elphic, R. C., Cayton, T. E., & McComas, D. J. (1998). The transport of plasma sheet material from the distant tail to geosynchronous orbit. Journal of Geophysical Research: Space Physics, 103(A9), 20297-20331. https://doi.org/10.1029/97JA03144

    Boyle, C. B., Reiff, P. H., & Hairston, M. R. (1997). Empirical polar cap potentials. Journal of Geophysical Research: Space Physics, 102(A1), 111-125. https://doi.org/10.1029/96JA01742

    Bruinsma, S. (2015). The DTM-2013 thermosphere model. J. Space Weather Space Clim., 5, A1. https://doi.org/10.1051/swsc/2015001

    Camporeale, E. (2019). The Challenge of Machine Learning in Space Weather: Nowcasting and Forecasting. Space Weather, 17(8), 1166-1207. https://doi.org/10.1029/2018SW002061

    Costello, K. A. (1998). Moving the Rice MSFM into a real-time forecast mode using solar wind driven forecast modules. Rice University.

    Crooker, N. U., & Gringauz, K. I. (1993). On the low correlation between long-term averages of solar wind speed and geomagnetic activity after 1976. Journal of Geophysical Research: Space Physics, 98(A1), 59-62. https://doi.org/10.1029/92JA01978

    Dungey, J. W. (1961). Interplanetary Magnetic Field and the Auroral Zones. Physical Review Letters, 6(2), 47-48. https://doi.org/10.1103/PhysRevLett.6.47

    Dutoit, S. H. C. (2012). Graphical exploratory data analysis. Place of publication not identified: Springer.

    Emery, B. A., Coumans, V., Evans, D. S., Germany, G. A., Greer, M. S., Holeman, E., Kadinsky-Cade, K., Rich, F. J., & Xu, W. (2008). Seasonal, Kp, solar wind, and solar flux variations in long-term single-pass satellite estimates of electron and ion auroral hemispheric power. Journal of Geophysical Research: Space Physics, 113(A6). https://doi.org/10.1029/2007JA012866

    Emmert, J. T., Drob, D. P., Picone, J. M., Siskind, D. E., Jones Jr., M., Mlynczak, M. G., Bernath, P. F., Chu, X., Doornbos, E., Funke, B., Goncharenko, L. P., Hervig, M. E., Schwartz, M. J., Sheese, P. E., Vargas, F., Williams, B. P., & Yuan, T. (2021). NRLMSIS 2.0: A Whole-Atmosphere Empirical Model of Temperature and Neutral Species Densities. Earth and Space Science, 8(3), e2020EA001321. https://doi.org/10.1029/2020EA001321

    Garrett, H. B., Dessler, A. J., & Hill, T. W. (1974). Influence of solar wind variability on geomagnetic activity. Journal of Geophysical Research (1896-1977), 79(31), 4603-4610. https://doi.org/10.1029/JA079i031p04603

    Haiducek, J. D., Welling, D. T., Ganushkina, N. Y., Morley, S. K., & Ozturk, D. S. (2017). SWMF Global Magnetosphere Simulations of January 2005: Geomagnetic Indices and Cross-Polar Cap Potential. Space Weather, 15(12), 1567-1587. https://doi.org/10.1002/2017SW001695

    Heidrich-Meisner, V., & Wimmer-Schweingruber, R. F. (2018). Chapter 16 - Solar Wind Classification Via k-Means Clustering Algorithm. In E. Camporeale, S. Wing, & J. R. Johnson (Eds.), Machine Learning Techniques for Space Weather (pp. 397-424). Elsevier. https://doi.org/10.1016/B978-0-12-811788-0.00016-0

    Hochreiter, S., & Schmidhuber, J. (1997). Long Short-Term Memory. Neural Comput., 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735

    Hoilijoki, S., Souza, V. M., Walsh, B. M., Janhunen, P., & Palmroth, M. (2014). Magnetopause reconnection and energy conversion as influenced by the dipole tilt and the IMF Bx. Journal of Geophysical Research: Space Physics, 119(6), 4484-4494. https://doi.org/10.1002/2013JA019693

    Ji, E.-Y., Moon, Y.-J., Park, J., Lee, J.-Y., & Lee, D.-H. (2013). Comparison of neural network and support vector machine methods for Kp forecasting. Journal of Geophysical Research: Space Physics, 118(8), 5109-5117. https://doi.org/10.1002/jgra.50500

    Johnson, J. R., & Wing, S. (2004). A cumulant-based analysis of nonlinear magnetospheric dynamics.

    Johnson, J. R., & Wing, S. (2005). A solar cycle dependence of nonlinearity in magnetospheric activity. Journal of Geophysical Research: Space Physics, 110(A4). https://doi.org/10.1029/2004JA010638

    Kivelson, M. G. R. C. T. (1995). Introduction to space physics. Cambridge University Press.

    Liemohn, M. W., McCollough, J. P., Jordanova, V. K., Ngwira, C. M., Morley, S. K., Cid, C., Tobiska, W. K., Wintoft, P., Ganushkina, N. Y., Welling, D. T., Bingham, S., Balikhin, M. A., Opgenoorth, H. J., Engel, M. A., Weigel, R. S., Singer, H. J., Buresova, D., Bruinsma, S., Zhelavskaya, I. S., . . . Vasile, R. (2018). Model Evaluation Guidelines for Geomagnetic Index Predictions. Space Weather, 16(12), 2079-2102. https://doi.org/10.1029/2018SW002067

    Liu, H.-L., Bardeen, C. G., Foster, B. T., Lauritzen, P., Liu, J., Lu, G., Marsh, D. R., Maute, A., McInerney, J. M., Pedatella, N. M., Qian, L., Richmond, A. D., Roble, R. G., Solomon, S. C., Vitt, F. M., & Wang, W. (2018). Development and Validation of the Whole Atmosphere Community Climate Model With Thermosphere and Ionosphere Extension (WACCM-X 2.0). Journal of Advances in Modeling Earth Systems, 10(2), 381-402. https://doi.org/10.1002/2017MS001232

    Mauk, B. H., & McIlwain, C. E. (1974). Correlation of Kp with the substorm-injected plasma boundary. Journal of Geophysical Research (1896-1977), 79(22), 3193-3196. https://doi.org/10.1029/JA079i022p03193

    McCulloch, W. S., & Pitts, W. (1943). A logical calculus of the ideas immanent in nervous activity. The bulletin of mathematical biophysics, 5(4), 115-133. https://doi.org/10.1007/BF02478259

    Newell, P. T., Sotirelis, T., Liou, K., Meng, C.-I., & Rich, F. J. (2007). A nearly universal solar wind-magnetosphere coupling function inferred from 10 magnetospheric state variables. Journal of Geophysical Research: Space Physics, 112(A1). https://doi.org/10.1029/2006JA012015

    Papitashvili, V. O., Papitashvili, N. E., & King, J. H. (2000). Solar cycle effects in planetary geomagnetic activity: Analysis of 36-year long OMNI dataset. Geophysical Research Letters, 27(17), 2797-2800. https://doi.org/10.1029/2000GL000064

    Qian, L., Burns, A. G., Emery, B. A., Foster, B., Lu, G., Maute, A., Richmond, A. D., Roble, R. G., Solomon, S. C., & Wang, W. (2014). The NCAR TIE-GCM. In Modeling the Ionosphere–Thermosphere System (pp. 73-83). https://doi.org/10.1002/9781118704417.ch7

    Schafer, J. L. (1999). Multiple imputation: a primer. Statistical methods in medical research, 8(1), 3-15.

    Shue, J.-H., Chao, J. K., Fu, H. C., Russell, C. T., Song, P., Khurana, K. K., & Singer, H. J. (1997). A new functional form to study the solar wind control of the magnetopause size and shape. Journal of Geophysical Research: Space Physics, 102(A5), 9497-9511. https://doi.org/10.1029/97JA00196

    Shue, J.-H., Song, P., Russell, C. T., Steinberg, J. T., Chao, J. K., Zastenker, G., Vaisberg, O. L., Kokubun, S., Singer, H. J., Detman, T. R., & Kawano, H. (1998). Magnetopause location under extreme solar wind conditions. Journal of Geophysical Research: Space Physics, 103(A8), 17691-17700. https://doi.org/10.1029/98JA01103

    Siebert, M. (Ed.). (1971). Maßzahlen der erdmagnetischen Aktivität. Springer.

    Skoug, R. M., Gosling, J. T., Steinberg, J. T., McComas, D. J., Smith, C. W., Ness, N. F., Hu, Q., & Burlaga, L. F. (2004). Extremely high speed solar wind: 29–30 October 2003. Journal of Geophysical Research: Space Physics, 109(A9). https://doi.org/10.1029/2004JA010494

    Subbotin, D. A., Shprits, Y. Y., & Ni, B. (2011). Long-term radiation belt simulation with the VERB 3-D code: Comparison with CRRES observations. Journal of Geophysical Research: Space Physics, 116(A12). https://doi.org/10.1029/2011JA017019

    Tabachnick, B. G., Fidell, L. S., & Ullman, J. B. (2007). Using multivariate statistics (Vol. 5). pearson Boston, MA.

    Tan, Y., Hu, Q., Wang, Z., & Zhong, Q. (2018). Geomagnetic Index Kp Forecasting With LSTM. Space Weather, 16(4), 406-416. https://doi.org/10.1002/2017SW001764

    Tenfjord, P., Østgaard, N., Strangeway, R., Haaland, S., Snekvik, K., Laundal, K. M., Reistad, J. P., & Milan, S. E. (2017). Magnetospheric response and reconfiguration times following IMF By reversals. Journal of Geophysical Research: Space Physics, 122(1), 417-431. https://doi.org/10.1002/2016JA023018

    Tsyganenko, N. A. (1989). A magnetospheric magnetic field model with a warped tail current sheet. Planetary and Space Science, 37(1), 5-20. https://doi.org/10.1016/0032-0633(89)90066-4

    Vaheb, H. (2020). Asset Price Forecasting using Recurrent Neural Networks. arXiv preprint arXiv:2010.06417.

    Wang, J., Zhong, Q., Liu, S., Miao, J., Liu, F., Li, Z., & Tang, W. (2015). Statistical analysis and verification of 3-hourly geomagnetic activity probability predictions. Space Weather, 13(12), 831-852. https://doi.org/10.1002/2015SW001251

    Wing, S., Johnson, J. R., Jen, J., Meng, C.-I., Sibeck, D. G., Bechtold, K., Freeman, J., Costello, K., Balikhin, M., & Takahashi, K. (2005). Kp forecast models. Journal of Geophysical Research: Space Physics, 110(A4). https://doi.org/10.1029/2004JA010500

    Wing, S., Newell, P. T., Sibeck, D. G., & Baker, K. B. (1995). A large statistical study of the entry of interplanetary magnetic field Y-component into the magnetosphere. Geophysical Research Letters, 22(16), 2083-2086. https://doi.org/10.1029/95GL02261

    Wintoft, P., Wik, M., Matzka, J., & Shprits, Y. (2017). Forecasting Kp from solar wind data: input parameter study using 3-hour averages and 3-hour range values. J. Space Weather Space Clim., 7, A29. https://doi.org/10.1051/swsc/2017027

    Xu, F., & Borovsky, J. E. (2015). A new four-plasma categorization scheme for the solar wind. Journal of Geophysical Research: Space Physics, 120(1), 70-100. https://doi.org/10.1002/2014JA020412

    Zhang, Y., & Paxton, L. J. (2008). An empirical Kp-dependent global auroral model based on TIMED/GUVI FUV data. Journal of Atmospheric and Solar-Terrestrial Physics, 70(8), 1231-1242. https://doi.org/10.1016/j.jastp.2008.03.008

    Zhelavskaya, I. S., Shprits, Y. Y., & Spasojević, M. (2017). Empirical Modeling of the Plasmasphere Dynamics Using Neural Networks. Journal of Geophysical Research: Space Physics, 122(11), 11,227-211,244. https://doi.org/10.1002/2017JA024406

    QR CODE
    :::