A Deep Learning Framework for Day Ahead Wind Power Short-Term Prediction

Due to the increasing proportion of wind power connected grid, day-ahead prediction plays a more and important role in operation system. This paper proposes short-term model based on deep learning (DWT_AE_BiLSTM). Firstly, discrete wavelet transform (DWT) is used denoise data, then an autoencoder (AE) technology extract data features, finally, bidirectional long memory (BiLSTM) for prediction. To verify effectiveness proposed DWT_AE_BiLSTM model, we studied three different stations compared their performance with shallow neural network model. Experimental analysis shows that this competitive forecasting accuracy stability. Compared BP has increased by 3.86%, 3.22% 3.42% farms, respectively.