Short-term wind power prediction method based on black kite algorithm and VMD

By combining the Blackwing Kite algorithm to optimize VMD parameters and constructing the TCN-BiGRU-Attention model, the problems of nonlinearity and spatiotemporal feature capture in wind power prediction were solved, achieving higher prediction accuracy and stability.

CN122246686APending Publication Date: 2026-06-19XINJIANG UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
XINJIANG UNIVERSITY
Filing Date
2026-03-20
Publication Date
2026-06-19

AI Technical Summary

Technical Problem

Existing technologies struggle to effectively capture the nonlinear and spatiotemporal characteristics of wind power, resulting in low accuracy in wind power prediction. Furthermore, traditional deep learning models suffer from insufficient generalization ability and excessive decomposition.

Method used

By combining the Black-winged Kite Algorithm (BKA), Variational Mode Decomposition (VMD), Temporal Convolutional Network (TCN), Bidirectional Gated Recurrent Unit (BiGRU), and Attention mechanism, an efficient short-term wind power prediction model is constructed. By optimizing VMD parameters and model hyperparameters, data processing capabilities are improved.

Benefits of technology

It significantly improves the accuracy of short-term wind power forecasting, reduces forecasting errors, and enhances the model's generalization ability and adaptability, especially performing well under different seasons and data conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122246686A_ABST
    Figure CN122246686A_ABST
Patent Text Reader

Abstract

This invention discloses a short-term wind power prediction method based on the Black-winged Kite Algorithm and Variational Mode Decomposition (VMD). The method includes the following steps: First, the parameters of Variational Mode Decomposition (VMD) are determined using the Black-winged Kite Algorithm (BKA), and VMD is used to decompose the data into several subsequences to reduce the complexity and instability of wind power time series. Second, each subsequence is combined with key meteorological data to form input components, and a combined prediction model using a Temporal Convolutional Network (TCN), a Bidirectional Gated Recurrent Unit (BiGRU), and an attention mechanism is used to predict each input component separately. Finally, the optimal combination of hyperparameters of the prediction model is determined using BKA, and the prediction results are summed and reconstructed to obtain the final wind power prediction result. The method provided by this invention further improves the prediction accuracy of short-term wind power, providing accurate wind power data for grid optimization and dispatching, thus ensuring the stable operation of wind power grid connection.
Need to check novelty before this filing date? Find Prior Art