利用EMD方法提取太阳活动周期成分

Extracting of Periodic Components in the Solar Activity Using EMD Method

EMD(经验模态分解)方法在处理非线性及非平稳时间序列时表现出了很大的优势和应用潜力．利用EMD方法研究太阳活动周期,对110年(1894—2003)和55年(1949—2003)的太阳黑子数月均值进行分解,分别得到一系列模式和一个趋势项,其中都可能包含有1．3至1．4年周期分量,25至30个月QBO(准双年振荡)分量,11年太阳周分量和22年Hale周分量．其中11年周期分量幅度最大,变化特征与太阳黑子数原始数据具有很高的相似性．不同于传统方法,EMD方法给出了太阳活动在不同时间尺度上各自分离的变化特征．

EMD （Empirical Mode Decomposition） technique has indicated the advantages and potentials in applications related to processing the nonlinear and non-stationary time series. In this paper, EMD method is adopted to study the solar cycles on various time scales. Using the EMD method, the averages of monthly medians of sunspot numbers of 110 years （1894-2003） and 55 years （1949-2003） can be decomposed into a series of modes and a trend respectively, including 1.3～1.4 years periodic component, 25～30 months QBO （Quasi-Biennial Oscillation） component, 11-year solar cycle component and 22 year Hale cycle component. Among those, the ll-year solar cycle mode with the largest amplitude has the high correlations with original sunspot numbers. In a nutshell, the EMD method other than the traditional and conventional methods can bring us the isolated characteristics of the solar variations on various time scales. Key words Solar cycle, Sunspot numbers, Empirical Mode Decomposition method, Intrinsic mode function