行星际日冕物质抛射地磁效应研究的支持向量机方法初步研究

叶煜东; 冯学尚

doi:10.11728/cjss2019.03.295

行星际日冕物质抛射地磁效应研究的支持向量机方法初步研究

doi: 10.11728/cjss2019.03.295

叶煜东^1,2,
冯学尚¹

1. 中国科学院国家空间科学中心空间天气学国家重点实验室北京 100190;
2. 中国科学院大学北京 100049

基金项目:

国家自然科学基金项目（41731067，41531073）和中国科学院"十三五"信息化建设专项（XXH13505-04）共同资助

详细信息

作者简介:
叶煜东,E-mail:ydye@spaceweather.ac.cn

中图分类号: P353
计量
- 文章访问数: 2865
- HTML全文浏览量: 62
- PDF下载量: 15575
- 被引次数: 0
出版历程
- 收稿日期: 2018-06-05
- 修回日期: 2019-01-31
- 刊出日期: 2019-05-15

Study on Geoeffectiveness of Interplanetary Coronal Mass Ejections by Support Vector Machine ormalsize

YE Yudong^1,2,
FENG Xueshang¹

1. State Key Laboratory of Space Weather, National Space Science Center, Chinese Academy of Sciences, Beijing 100190;
2. University of Chinese Academy of Sciences, Beijing 100049

摘要

摘要: 行星际日冕物质抛射（Interplanetary Coronal Mass Ejection，ICME）与地球磁层相互作用并带来地磁暴等地磁扰动.从Richardson和Cane提供的近地球ICME列表中筛选出ICME事件集，基于ICME扰动期间的行星际等离子体与磁场数据提取出特征.通过计算各特征的费舍尔分值（Fisher Score），对这些特征进行选择，发现行星际磁场南北向分量持续时间小于-10nT且激波等扰动所带来的ICME扰动开始时，太阳风速度的增量等特征与ICME事件的地磁效应密切相关.这与现有的传统统计研究结果一致.以这些特征为基础，训练得到的径向基函数支持向量机能够以0.78±0.08的准确率判断ICME事件是否会产生中等及以上强度的地磁暴（Dst ≤-50nT）.
- 支持向量机 /
- 行星际日冕物质抛射 /
- 地磁效应
Abstract: As arriving at the Earth, Interplanetary Coronal Mass Ejections (ICME) will interact with the Earth's magnetosphere and cause geomagnetic storms. The ICME event set is obtained by Richardson and Cane's Near Earth ICME list, and the input features are extracted based on interplanetary solar wind and magnetic data during ICME disturbance. A total of 483 ICME events from 1996 to 2006 are chosen in this study. 13 magnetic and kinetic features are finally selected for the training of the machine learning model. Rank of each feature's Fisher score indicates that the duration of the south-directed interplanetary magnetic field that is larger than 10nT and the increase of solar wind speed at the upstream shock or wave disturbance is closely related to the geoeffectiveness of ICME events, which is consistent with those former statistical results. The trained Radial Basis Function Support Vector Machine (RBF-SVM) can determine whether an ICME event could trigger moderate or stronger geomagnetic storms (Dst ≤ -50nT) effectively with an accuracy of 0.78±0.08. The results show that RBF-SVM can be used as a powerful tool in further analysis, and the better prediction of the geoeffectiveness of ICME will be obtained.
- Support Vector Machine /
- Interplanetary Coronal Mass Ejections /
- Geo-effectiveness

HTML全文

参考文献(25)

[1]	ZURBUCHEN T H, RICHARDSON I G. In-situ solar wind and magnetic field signatures of interplanetary coronal mass ejections[J]. Space Sci. Rev., 2006, 123(1/2/3):31-43
[2]	LAKHINA G S, TSURUTANI B T. Geomagnetic storms:historical perspective to modern view[J]. Geosci. Lett., 2016, 3(1):5
[3]	GONZALEZ W D, JOSELYN J A, KAMIDE Y, et al. What is a geomagnetic storm[J]. J. Geophys. Res., 1994, 99:5771-5792
[4]	WU J G, LUNDSTEDT H. Geomagnetic storm predictions from solar wind data with the use of dynamic neural networks[J]. J. Geophys. Res.:Space Phys., 1997, 102(A7):14255-14268
[5]	BALA R, REIFF P. Improvements in short-term forecasting of geomagnetic activity[J]. Space Weather, 2012, 10(6).DOI: 10.1029/2012SW000779.
[6]	WATANABE S, SAGAWA E, OHTAKA K, et al. Prediction of the Dst index from solar wind parameters by a neural network method[J]. Earth Planets Space, 2014, 54(12):1263-1275
[7]	WANG Y M, YE P Z, WANG S, et al. A statistical study on the geoeffectiveness of Earth-directed coronal mass ejections from March 1997 to December 2000[J]. J. Geophys. Res., 2002, 107(A11).DOI: 10.1029/2002JA009244.
[8]	GOPALSWAMY N, YASHIRO S, XIE H, et al. Properties and geoeffectiveness of magnetic clouds during solar cycles 23 and 24[J]. J. Geophys. Res.:Space Phys., 2015, 120(11):9221-9245
[9]	LAWRANCE M B, SHANMUGARAJU A, MOON Y J, et al. Relationships between interplanetary coronal mass ejection characteristics and geoeffectiveness in the rising phase of solar cycles 23 and 24[J]. Sol. Phys., 2016, 291(5):1547-1560
[10]	CANE H V, RICHARDSON I G. Interplanetary coronal mass ejections in the near-Earth solar wind during 1996-2002[J]. J. Geophys. Res., 2003, 108(A4).DOI:10. 1029/2002JA009817.
[11]	RICHARDSON I G, CANE H V. Near-earth interplanetary coronal mass ejections during solar cycle 23(1996-2009):catalog and summary of properties[J]. Sol. Phys., 2010, 264(1):189-237
[12]	MATHEWS G J, TOWHEED S S. NSSDC OMNIWeb:The first space physics WWW-based data browsing and retrieval system[J]. Comp. Networks ISDN Syst., 1995, 27(6):801-808
[13]	GU Q Q, LI Z H, HAN J W. Generalized Fisher Score for Feature Selection[C]//Proceedings of the 27th Conference on Uncertainty in Artificial Intelligence, 2011:266-273
[14]	PEDREGOSA F, VAROQUAUX G, GRAMFORT A, et al. Scikit-learn:machine learning in Python[J]. J. Mach. Learn. Res., 2011, 12:2825-2530
[15]	CHI Yutian, SHEN Chenglong, WANG Yuming, et al. Statistical study of the interplanetary coronal mass ejections from 1995 to 2015[J]. Sol. Phys., 2016, 291(8):2419-2439
[16]	WANG Yuming, YE Pinzhong, WANG Shui. An interplanetary origin of great geomagnetic storms:multiple magnetic clouds[J]. Chin. J. Geophys., 2004, 47(3):417-423
[17]	TSURUTANI B T, GONZALEZ W D, TANG F, et al. Great magnetic storms[J]. Geophys. Res. Lett., 1992, 19(1):73-76
[18]	BURTON R K, MCPHERRON R L, RUSSELL C T. An empirical relationship between interplanetary conditions and Dst[J]. J. Geophys. Res., 1975, 80(31):4204-4214
[19]	VAPNIK V, GOLOWICH S E, SMOLA A J. Support vector method for function approximation, regression estimation and signal processing[C]//Proceedings of the Advances in Neural Information Processing Systems, 1997:281-287
[20]	AMARI S, MURATA N, MULLER K R, et al. Asymptotic statistical theory of overtraining and cross-validation[J]. IEEE Trans. Neural Netw., 1997, 8(5):985-996
[21]	REFAEILZADEH P, TANG L, LIU H. Cross-validation[R]. Encyclopedia of Database System, US:Springer. DOI: 10.1007/978-0-387-39940-9_565
[22]	KOHAVI R. A study of cross-validation and bootstrap for accuracy estimation and model selection[C]//Proceedings of the 14th International Joint Conference on Artificial Intelligence-Volume 2. Quebec:Morgan Kaufmann Publishers Inc, 1995:1137-1143
[23]	FLUECK J. A study of some measures of forecast verification[C]//Proceedings of the Preprints, 10th Conference on Probability and Statistics in Atmospheric Sciences. Edmonton:Alberta, Am. Meteor. Soc., 1987:69-73
[24]	SHEN Fang, WANG Yuming, SHEN Chenglong, et al. On the collision nature of two coronal mass ejections:a review[J]. Sol. Phys., 2017, 292(8):104
[25]	SHEN Chenglong, CHI Yutian, WANG Yuming, et al. Statistical comparison of the ICME's geoeffectiveness of different types and different solar phases from 1995 to 2014[J]. J. Geophys. Res.:Space Phys., 2017, 122(6):5931-5948