Multiple Time-scale Characteristics Analysis of Rainfall in Hunan Province Based on Ensemble Empirical Mode Decomposition
Guoyong Zhang, Bo Li, Li Li, Xiudong Zhou, Xunjian Xu, Lifu He · 2019
In view of the mode mixing and end effects of empirical mode decomposition (EMD), the ensemble empirical mode decomposition (EEMD) method based on extreme learning machine (ELM) signal continuation is presented to analyze the rainfall time series with multiple time-scale, In this paper, the EEMD and the wavelet method were applied to analyze the annual rainfall sequence in Hunan province. The results show that the EEMD method, as a kind of new signal processing method, can obtain accurate characteristics of the annual rainfall series. Based on this, it can be found that the proposed method can be widely used for the multiple timescale characteristics analysis of rainfall time series.