Construction of Natural Gas Price Prediction Model Based on EEMD-GRU Neural Network
In view of the low accuracy of traditional methods in predicting natural gas prices,this paper adopts a gating cycle mechanism based on collective empirical mode decomposition to predict natural gas prices.The collective empirical mode decom-position method is used to adaptively decompose the natural gas price.The intrinsic mode function components that need high frequency are selected through the autocorrelation method.Several high frequency components with low autocorrelation coeffi-cients are decomposed.The characteristics of the decomposed high-frequency intrinsic mode function components are analyzed by the gated loop mechanism neural network.They are reconstructed and combined with the low-frequency intrinsic mode func-tion components,which makes better use of the useful information in the natural gas price data set.The experimental results show that this method is more accurate and robust than other traditional methods in predicting natural gas prices.
set empirical mode decompositiongating mechanismnatural gasprice prediction
万方数据
维普
