Short-term Wind Speed Prediction Based on RD-DW-BiLSTM

Aiming at the problem of future information leakage in traditional hybrid methods in wind speed prediction, a Rolling Decomposition data preprocessing method is proposed. By observing data characteristics, a fixed window length is determined, and the data within the window is subjected to real-time variational mode decomposition. The decomposition results are then input into the BiLSTM model for prediction. Additionally, a Dynamic Weighting method is introduced to assign different weights to the prediction results at different time points. Experimental results show that the RD-DW-BiLSTM model can effectively mitigate future information leakage while maintaining good predictive accuracy, enabling accurate wind speed predictions.