Multi-dimensional wind power prediction based on time-series characterization analysis and VMD-EMD quadratic decomposition

Wind power data receive wind volatility and have strong non-smoothness, making it difficult to achieve high accuracy in wind power prediction. To address this challenge, this paper proposes a wind power multi-step prediction model combining VMD (Variational Modal Decomposition), EMD (Empirical Modal Modal Decomposition), Hurst analysis and temporal entropy values. Firstly, the first decomposition of the historical wind power data is carried out by VMD decomposition ; then Hurst analysis is performed on the components of the first decomposition, and the components with low regularity are decomposed twice using EMD decomposition; then the components of the second decomposition are further filtered using permutation entropy, and the components with high entropy values are compared with the high The components with high entropy, high Hurst and low entropy of the secondary decomposition are formed into high randomness irregular component, regular component and low randomness low regularity component; for the three types of components, BP neural network is used to predict them respectively, and they are reorganized into wind power prediction values. The experiments prove that the model proposed in this paper has higher prediction accuracy and faster running time than the current mainstream models, and can achieve more efficient wind power prediction.