Multi-step ahead wind speed forecasting approach coupling maximal overlap discrete wavelet transform, improved grey wolf optimization algorithm and long short-term memory

Abstract

Accurate and reliable wind speed forecasting is of great significance to the management and utilization of wind energy. An improved deep learning model for wind speed forecasting, abbreviated as MODWT-RF-IGWO-LSTM, is presented in this paper. Firstly, the maximum overlap discrete wavelet transform (MODWT) is applied to denoise the original wind speed series. Secondly, the random forest (RF) algorithm is used for feature selection. Thirdly, the improved grey wolf optimization algorithm (IGWO) is applied to optimize the parameters of the long short-term memory (LSTM) model. Finally, the denoised wind speed data is entered into the well-trained LSTM model to obtain the final wind speed forecasting result. The performance of the proposed model is assessed by actual wind speed data for three different months of the year. The experimental results show that the proposed deep learning model for wind speed forecasting has good predictive ability. And the proposed model performs better than other benchmark models in this paper.