摘要
针对地磁变化场时间序列的混沌特性,提出了一种改进的集成经验模态分解(modified ensemble empirical mode decomposition,MEEMD)-样本熵-最小二乘支持向量机(least square support vector machine,LSSVM)的地磁变化场预测模型。首先,利用MEEMD-样本熵将非平稳的地磁变化场时间序列分解为一系列复杂度差异明显的地磁变化场子序列;然后,针对每一个子序列分别建立LSSVM模型,选择各自适合的最优模型参数;最后,以地磁台站实测的地磁变化场数据为例进行实验,并与基于单一LSSVM以及RBF径向基神经网络的两种预测模型进行比较。实验结果表明,MEEMD-样本熵-LSSVM模型的预测值能紧跟地磁变化场的变化趋势,相比另外两种模型,体现出更好的预测效果,在地磁Kp指数小于3时,预测3h平均绝对误差为1.63nT。
Modeling and forecasting of the geomagnetic variation field is the important research topic of geomagnetic navigation and space environment monitoring.According to the chaotic feature of geomagnetic variation time series,a combined forecasting model based on modified ensemble empirical mode decomposition(MEEMD)-sample entropy(SampEn)-least square support vector machine(LSSVM)is proposed.Firstly,the geomagnetic variation time series is decomposed into a series of geomagnetic variation subsequences with obvious differences in complex degree using MEEMD-SampEn. Then,the forecasting model of each subsequence is created with LSSVM using the optimal model parameters.Finally,the simulation is performed by using the real data collected from the geomagnetic observatory.The results show that the forecasting value of the MEEMD-SampEn-LSSVM model can closely keep up with the trend of geomagnetic variation field,and obviously better than the other two models.The mean absolute error of the model forecasting three hours is 1.63nT when Kpless than 3.
引文
[1]Yi Shihua,Liu Daizhi,He Yuanlei,et al.Modeling and Forecasting of the Variable Geomagnetic Field by Support Vector Machine[J].Chinese Journal of Geophysics,2013,56(1):127-135(易世华,刘代志,何元磊,等.变化地磁场预测的支持向量机建模[J].地球物理学报,2013,56(1):127-135)
[2]Xu Wenyao.Physics of Electromagnetic Phenomena of the Earth[M].Hefei:Press of University of Science and Technology of China,2009:473-474(徐文耀.地球电磁现象物理学[M].合肥:中国科学技术大学出版社,2009:473-474)
[3]Xu Wenyao.Improvement of Scaling and Evaluating of K Index[J].Northwestern Seismological Journal,2005,27(supp.):36-41(徐文耀.地磁活动K指数值量算和确定方法的改进[J].西北地震学报,2005,27(增):36-41)
[4]Peng Fuqing.Geomagnetic Model and Geomagnetic Navigation[J].Hydrographic Surveying and Charting,2006,26(2):73-75(彭富清.地磁模型与地磁导航[J].海洋测绘,2006,26(2):73-75)
[5]Zheng Hui,Wang Yong,Wang Hubiao,et al.Simulation Research of Earth’s Gravity and Geomagnetism Potential Field Aided Underwater Navigation[J].Geomatics and Information Science of Wuhan University,2012,37(10):1 198-1 202(郑晖,王勇,王虎彪,等.地球重磁位场辅助水下潜艇导航方针研究[J].武汉大学学报·信息科学版,2012,37(10):1 198-1 202)
[6]Zhao Jianhu,Wang Shengping,Wang Aixue,et al.Study on the Selection of the Geomagnetic Adaptable Matching Area Based on the Geomagnetic Co-occurrence Matrix[J].Geomatics and Information Science of Wuhan University,2011,36(4):446-449(赵建虎,王胜平,王爱学,等.基于地磁共生矩阵的水下地磁导航适配区选择[J].武汉大学学报·信息科学版,2011,36(4):446-449)
[7]Niu Chao,Li Xihai,Liu Daizhi.Chaotic Dynamic Characteristics of Z Component in Geomagnetic Variation Field[J].Acta Physica Sinica,2010,59(5):3077-3087(牛超,李夕海,刘代志.地磁变化场Z分量的混沌动力学特性分析[J].物理学报,2010,59(5):3 077-3 087)
[8]Zheng Xu,Hao Zhiyong,Jin Yang,et al.Studying Noise Contributions of IC Engine via MEEMD Method[J].Journal of Zhejiang University(Engineering Science),2012,46(5):954-960(郑旭,郝志勇,金阳,等.基于MEEMD的内燃机辐射噪声贡献[J].浙江大学学报(工学版),2012,46(5):954-960)
[9]Richman J S,Moorman J R.Physiological Time-series Analysis Using Approximate Entropy and Sample Entropy[J].Am J Physiol Heart Circ Physiol,2000,278:2 039-2 049
[10]Suykens J A K,Gestel T V,Brahanter J D,et al.Least Squares Support Vector Machines[J].River Edge:World Scientific,2002:71-148
[11]Qi Wei,Wang Xiufang,Li Xihai,et al.Selection of Characteristic Components for Geomagnetic Matching Based on Statistical Modeling[J].Progress in Geophysics,2010,25(1):324-330(齐玮,王秀芳,李夕海,等.基于统计建模的地磁匹配特征量选择[J].地球物理学进展,2010,25(1):324-330)
[12]Chen G X,Xu W Y,Du A M,et al.Statistical Characteristics of the Day-to-Day Variability in the Geomagnetic Sq field[J].J Geophys Res,2007,112:320-330