Wind speed forecasting using a new multi-factor fusion and multi-resolution ensemble model with real-time decomposition and adaptive error correction. (1st August 2020)

• Record Type: Journal Article
• Authors: Liu, Hui; Yang, Rui; Duan, Zhu
• Is Part Of: Energy conversion and management. Volume 217(2020)
• Publication Date: 2020-08-01
• Subjects: NWP Numerical Weather Prediction -- MAE Mean Average Error -- WRF Weather Research and Forecasting -- MAPE Mean Absolute Percentage Error -- MM5 Fifth-generation Mesoscale Model -- RMSE Root Mean Square Error -- ELM Extreme Learning Machine -- LBQ-test Ljung-Box Q-Test -- MLP Multi-Layer Perceptron -- ASREM Average/Staked auto encoder Resolution Ensemble Model -- ENN Elman Neural Network -- CASREM Corrected Average/Staked auto encoder Resolution Ensemble Model -- LSTM Long Short-Term Memory -- CMPASREM Corrected Multi-Position Average/Staked auto encoder Resolution Ensemble Model -- ARMA Auto-Regression Moving Averaging -- A-BiLSTM Average-Bidirectional Long Short-Term Memory -- BiLSTM Bidirectional Long Short-Term Memory -- S-BiLSTM Staked auto encoder- Bidirectional Long Short-Term Memory -- B-GRUNNs Bidirectional Gated Recurrent Unit Neural Networks -- AFSA-ACO Artificial Fish Swarm Algorithm and Ant Colony Optimization -- GARCH Generalized Auto-Regressive Conditionally Heteroscedastic -- KPCA Kernel principal component analysis -- LSSVM Least Squares Support Vector Machine -- FPCA Functional Principal Component Analysis -- ORELM Outlier-Robust Extreme Learning Machine -- AE Auto-Encoder -- VMD Variational Mode Decomposition -- SAE Staked Auto-Encoder -- ICEEMDAN Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise -- WCA Water Cycle Algorithm -- MODWPT Maximal Overlap Discrete Wavelet Packet Transform -- MVO Multi-universe Optimization -- NSGA-III Non-dominated Sorting Genetic Algorithm III -- MOGWO Multi-Objective Grey Wolf Optimizer -- MOGOA Multi-Objective Grasshopper Optimization Algorithm -- MOICA Multi-Objective Imperialist Competitive Algorithm; Wind speed forecasting -- Multi-resolution ensemble -- Adaptive multiple error corrections -- Multi-position data fusion -- Real-time decomposition; 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.2020.112995
• Languages: English
• ISSNs: 0196-8904

Abstract

Highlights: Multi-resolution ensemble can make full use of hidden representation in the dataset. Real-time decomposition can improve the feasibility of engineering applications. Adaptive multiple error corrections enable the secondary development of predictable components in residual errors. Multi-position data fusion can combine correlations of multi-position information. Abstract: The chaos and non-stability of the wind pose challenges for obtaining reliable wind speed forecasting. Accurate and reliable wind speed forecasting is critical to the efficient and safe operation of wind power systems. In the study, a new hybrid multi-factor fusion and multi-resolution ensemble model is proposed for forecasting wind speed. The proposed hybrid forecasting model adopts real-time data decomposition and adaptive multiple error correction strategies and it consists of three phases. In stage I, the low-resolution and high-resolution data obtained by the averaging method and the Staked Auto-Encoder (SAE) feature extraction method, respectively. The data are used by Bidirectional Long Short-Term Memory (BiLSTM) for multi-step forecasting. The forecasting results are reasonably ensemble by Non-dominated Sorting Genetic Algorithm III (NSGA-III) to complete the multi-resolution ensemble phase. In stage II, Ljung-Box Q-Test (LBQ-test) is utilized to detect predictable components in the forecasting results. And Auto-Regression Moving Averaging (ARMA) completes adaptive multiple error corrections. In stage III, Multi-universe Optimization (MVO) is used to ensemble the wind speed forecasting results of multiple positions. At this point, the multi-position data fusion phase is completed. Wind speed data from Xinjiang, China are carried out to validate the efficiency of the proposed model. Experimental analysis shows: The performance of the proposed hybrid model is superior to other comparative models. The MAE error of 3-step wind speed forecasting from the four sets of experiments is only 0.4614 m/s, 0.4482 m/s, 0.6755 m/s, and 0.2912 m/s, respectively.