Multi-step wind speed forecasting based on hybrid multi-stage decomposition model and long short-term memory neural network

• Published in: Energy conversion and management Vol. 213; p. 112869
• Main Authors: Rodrigues Moreno, Sinvaldo, Gomes da Silva, Ramon, Cocco Mariani, Viviana, dos Santos Coelho, Leandro
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.06.2020; Elsevier Science Ltd
• Subjects: Adaptive systems; Artificial neural network; Artificial neural networks; Decomposition; Decomposition based methods; Forecasting; Fuzzy logic; Gaussian process; Learning; Long short-term memory; Mathematical models; Neural networks; Regression analysis; Spectral analysis; Spectrum analysis; Spectrum singular analysis; Support vector machines; Variational mode decomposition; Wind; Wind speed; Long short-term memory; Variational mode decomposition; Spectrum singular analysis; Artificial neural network; Wind speed; Decomposition based methods
• ISSN: 0196-8904; 1879-2227
• DOI: 10.1016/j.enconman.2020.112869

[Display omitted] •New hybrid decomposition and an effective ensemble learning model for wind speed time series forecasting is proposed.•The AM-FM theory combined with an ensemble of VMD and SSA-LSTM is an original and powerful approach presented in this paper to wind speed forecasting.•Introducing the ensemble-learning model based on VMD and SSA, showing that it acts as an effective way for achieving reliable and stable forecasting. The intermittent nature of wind can represent an obstacle to get reliable wind speed forecasting, thus many methods were developed to improve the accuracy, due to unstable behavior patterns and the presence of noise signal. In order to overcome this issue, a preprocessing step is desirable to provide more reliable data. Decomposition strategy is reported as the crucial component of this improving task of the wind speed forecasting. It can be applied as the first step or as a recurrent process, and normally the raw wind speed data is decomposed in several signal patterns. Based on this understanding, this paper proposed a combination of two signal decomposition strategies, known as variational mode decomposition (VMD) and singular spectral analysis (SSA), with modulation signal theory. The proposed decomposition approach is further coupled with a long short-term memory neural network (LSTM), the adaptive neuro-fuzzy system (ANFIS), echo state network (ESN), support vector regression (SVR) and Gaussian regression process (GRP) models resulting in new ensemble learning approaches. All results obtained through these ensembles are compared between them and demonstrated an error stabilization behavior, ability decomposing the wind speed into uncorrelated components, reducing the errors from one up to twelve steps-ahead forecasting. In general terms, the results indicate that ensembles learning framework are robust and reliable to applications in wind speed forecasting task.