A Hybrid Model Based on a Two-Layer Decomposition Approach and an Optimized Neural Network for Chaotic Time Series Prediction

摘要: The prediction of chaotic time series has been a popular research field in recent years. Due to the strong non-stationary and high complexity series, it is difficult directly analyze predict depending on single model, so hybrid model become promising favorable alternative. In this paper, we put forward novel based two-layer decomposition approach an optimized back propagation neural network (BPNN). proposed obtain comprehensive information which composed complete ensemble empirical mode with adaptive noise (CEEMDAN) variational (VMD). VMD algorithm used for further frequency subsequences obtained by CEEMDAN, after performance significantly improved. We then use BPNN firefly (FA) prediction. experimental results indicate that superior other competing approaches terms four evaluation indexes one-step multi-step ahead predictions. good prospect series.