摘要
利用EMD方法对海浪观测资料进行处理 ,通过在数据两端的“平衡位置”处分别附加平行直线段的方法进行端点抑制 ,分解出 1 0个内在模函数和 1个剩余趋势项 ,再对各内在模函数进行Hilbert变换 ,得到波浪的Hilbert谱。对所得结果的分析表明 ,各模态在Hilbert谱中的分布趋势和Fourier谱中谱线的变化趋势是一致的 ,第一模态的中心频率与Fourier谱的谱峰频率相对应 ;EMD方法是对非线性、非平稳过程数据进行距平化的好方法 ,距平化的过程和消除趋势项的处理是统一的。
The empirical mode decomposition (EMD) method is used to process the observed wave data, constant values are added respectively at the balanced positions of both data ends to restrain the end effects, and the wave data are then decomposed into ten intrinsic mode functions (IMFs) and a residual trend term. Hilbert transform is made for each of ten IMFs to obtain Hilbert spectrum of wave. It is shown from the analysis results that the distributing trend of IMFs in Hilbert spectrum is consistent with the varying trend of spectral lines in Fourier spectrum, and the central frequency of the first IMF corresponds to the spectral peak frequency of Fourier spectrum. Therefore, it is suggested that the EMD method is a good method for demeaning the nonlinear and nonstationary process data, and the demeaning process and the data detrending are unified.
出处
《黄渤海海洋》
CSCD
北大核心
2002年第2期12-21,共10页
Journal of Oceanography of Huanghai & Bohai Seas
基金
国家海洋局青年海洋科学基金资助项目 (982 0 4)
