Computer Engineering and Applications ›› 2017, Vol. 53 ›› Issue (24): 252-256.DOI: 10.3778/j.issn.1002-8331.1606-0395

Previous Articles     Next Articles

EEMD and RBF neural network prediction of sunspot monthly mean

SUN Tangle, LI Guohui   

  1. School of Electronic Engineering, Xi’an University of Posts and Telecommunications, Xi’an 710121, China
  • Online:2017-12-15 Published:2018-01-09



  1. 西安邮电大学 电子工程学院,西安 710121

Abstract: The sunspot monthly mean is a typical chaotic time series. It has strong nonlinear and non-stationary characteristics, and can reflect the true level of the solar activity. A forecasting model of combinating Ensemble Empirical Mode Decomposition(EEMD) with Radial Basis Function(RBF) neural network is adopted. The original time series is decomposed into a number of different time scales intrinsic mode function by using EEMD, and then these components are modeled and predicted. The predicted value of the original time series is reconstructed by the predictive value of each component. The model not only reduces the complexity of the algorithm, but also improves the physical meaning of the modal components. The simulation results show that compared with the Empirical Modal Decomposition(EMD) and RBF combination model, the model has higher prediction accuracy.

Key words: sunspot, ensemble empirical mode decomposition, Radial Basis Function(RBF) neural network, prediction

摘要: 太阳黑子月均值是典型的混沌时间序列,具有较强的非线性和非平稳特征,能够反映太阳活动的真实水平。采用一种应用集合经验模态分解(Ensemble Empirical Mode Decomposition,EEMD)与径向基函数(Radial Basis Function,RBF)神经网络组合的预测模型。通过EEMD将原始时间序列分解为若干个不同时间尺度的本征模态函数(Intrinsic Mode Function,IMF)分量,并对这些分量进行建模预测,再将各分量的预测值重构得到原始时间序列的预测值,这样不仅降低了算法的复杂性,而且有利于提高模态分量包含信息的物理意义。仿真结果表明,与经验模态分解(Empirical Mode Decomposition,EMD)结合RBF神经网络的模型相比,该模型具有较高的预测精度。

关键词: 太阳黑子, 集合经验模态分解, 径向基函数(RBF)神经网络, 预测