Big multi-step wind speed forecasting model based on secondary decomposition, ensemble method and error correction algorithm. (15th January 2018)

• Record Type: Journal Article
• Authors: Liu, Hui; Duan, Zhu; Han, Feng-ze; Li, Yan-fei
• Is Part Of: Energy conversion and management. Volume 156(2018)
• Page Start: 525
• Page End: 541
• Publication Date: 2018-01-15
• Subjects: NWP numerical weather prediction -- CS Cuckoo search -- FS fuzzy system -- WRF weather research and forecasting -- KF Kalman filter -- ARIMA auto-regressive integrated moving average -- ARCH autoregressive conditional heteroskedasticity -- ANN artificial neural networks -- SVM support vector machine -- CRO Coral Reefs optimization algorithm -- ELM extreme learning machine -- MFNN multi-layer feed-forward neural network -- SPSA simultaneous perturbation stochastic approximation -- HM Hammerstein Model -- AR auto-regressive -- AdaBoost adaptive boosting -- MLP multilayer perceptron -- DNN-MRT deep neural network based meta regression and transfer learning -- WD wavelet decomposition -- FEEMD fast ensemble empirical mode decomposition -- EMD empirical mode decomposition -- WPD wavelet packet decomposition -- SSA singular spectrum analysis -- BFGS Broyden–Fletcher–Goldfarb–Shanno Quasi-Newton Back Propagation -- LSSVM least square support vector machine -- PSOGSA partial swarm optimization combined with gravitational search algorithm -- FCM fuzzy C-means -- EEMD ensemble empirical mode decomposition -- SampEn sample entropy -- VMD variational mode decomposition -- MAdaBoost Modified AdaBoost.RT -- WF wavelet filter -- MAE mean absolute error -- MAPE mean absolute percentage error -- RMSE root mean squared error -- CWT continuous wavelet transform -- DWT discrete wavelet transform -- LMD local mean decomposition -- ADMM alternate direction method of multipliers; Big multi-step wind speed forecasting -- Wavelet decomposition -- Variational mode decomposition -- Sample entropy -- Modified adaBoost.RT -- Wavelet filter; Direct energy conversion -- Periodicals; Energy storage -- Periodicals; Energy transfer -- Periodicals; Énergie -- Conversion directe -- Périodiques; Direct energy conversion; Periodicals; 621.3105
• DOI: 10.1016/j.enconman.2017.11.049
• Languages: English
• ISSNs: 0196-8904

Abstract

Highlights: A new hybrid method is proposed for the wind speed multi-step forecasting. The wavelet decomposition is adopted to reduce the noise of the original data. The different decomposing algorithms are used to decompose the original data. The different forecasting models are built to predict the pre-processed data. The error correction method is proposed to correct the wrong predictions. Abstract: Wind power is one of the most promising powers. Wind speed forecasting can eliminate the harmful effect caused by the intermittent and fluctuation of wind power, and big multi-step forecasting can provide more time for the power grid to be adjusted. To achieve the high-precision big multi-step forecasting, a novel hybrid model named as the WD-SampEn-VMD-MadaBoost-BFGS-WF is proposed in the study, which consisting of three main modeling steps including the secondary decomposition, the ensemble method and the error correction. The detail of the proposed model is given as follows: (a) wind speed series are decomposed by the WD ( Wavelet Decomposition ) to obtain wind speed subseries. The SampEn ( Sample Entropy ) algorithm is used to estimate the unpredictability of these wind speed subseries. The most unpredictable subseries will be decomposed secondarily by the VMD ( Variational Mode Decomposition ); (b) the subseries are proceeded by the MAdaBoost ( Modified AdaBoost.RT ) with the BFGS ( Broyden–Fletcher–Goldfarb–Shanno Quasi-Newton Back Propagation ) neuron network to obtain forecasting subseries; (c) all of the forecasting subseries will be combined with the original subseries to form the combined wind speed series, which will be further proceeded by the WF ( Wavelet Filter ) to obtain the corrected forecasting series from the point of the frequency domain; (d) the corrected forecasting series are reconstructed to get the final forecasting series. To validate the effectiveness of the proposed model, several forecasting cases are provided in the study. The result indicates that the proposed model has satisfactory forecasting performance in the big multi-step extremely strong simulating wind speed forecasting.