Application of hybrid model based on empirical mode decomposition, novel recurrent neural networks and the ARIMA to wind speed prediction

• Published in: Energy conversion and management Vol. 233; p. 113917
• Main Authors: Liu, Ming-De, Ding, Lin, Bai, Yu-Long
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.04.2021; Elsevier Science Ltd
• Subjects: administrative management; Autoregressive moving-average models; China; Complexity; Datasets; Decomposition; Electric power generation; Empirical mode decomposition; energy conversion; Entropy; Hybrid system; Long short-term memory neural network; Neural networks; power generation; prediction; Prediction models; Recurrent neural networks; Sequences; Wind power; Wind power generation; Wind speed; Wind speed prediction; China; Empirical mode decomposition; Recurrent neural networks; Long short-term memory neural network; Wind speed prediction; Hybrid system

•EMD is applied to reduce the complexity of the prediction subsequence.•The SE of the decomposed sub-sequence is induced to improve the prediction precision of wind speed.•LSTM is applied to predict the high frequency sub-sequences.•The ARIMA is employed to predict the low frequency sub-sequences an...