Wind Speed Forecasting Based on a Hybrid EMD-BLS Method

Wind power is a kind of environmentally friendly and economical renewable natural energy resources. Exact speed of wind forecasting act an important part in wind power planning. However, the time series (TS) of wind speed is non-linear and non-stationary, which make it difficult to be forecasted. A novelty forecasting method of one-step ahead wind speed which based on empirical mode decomposition (EMD) and broad learning system (BLS) methods is proposed. EMD is a powerful tool for nonlinear data decomposition in such a noisy environment and the purpose is to decompose a complex TS into different frequency components. The components after decomposing can reduce the non-stationary of TS. Subsequently, a newly developed network framework called BLS provided another way to learn deep structures and is trained to forecasting each component. The final prediction result is obtained from superposition of the prediction value of each component. Moreover, the effectiveness of the hybrid EMD-BLS method is verified through two real-world wind speed data sets.

[1]  Haibin Yu,et al.  Day-ahead hourly photovoltaic generation forecasting using extreme learning machine , 2015, 2015 IEEE International Conference on Cyber Technology in Automation, Control, and Intelligent Systems (CYBER).

[2]  Henrik Madsen,et al.  Conditional Weighted Combination of Wind Power Forecasts , 2010 .

[3]  Wang Xiaolan,et al.  One-Month Ahead Prediction of Wind Speed and Output Power Based on EMD and LSSVM , 2009, 2009 International Conference on Energy and Environment Technology.

[4]  Chee Kheong Siew,et al.  Extreme learning machine: Theory and applications , 2006, Neurocomputing.

[5]  C. L. Philip Chen,et al.  Broad Learning System: An Effective and Efficient Incremental Learning System Without the Need for Deep Architecture , 2018, IEEE Transactions on Neural Networks and Learning Systems.

[6]  N. Huang,et al.  The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis , 1998, Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences.

[7]  C. L. Philip Chen,et al.  Structured Manifold Broad Learning System: A Manifold Perspective for Large-Scale Chaotic Time Series Analysis and Prediction , 2019, IEEE Transactions on Knowledge and Data Engineering.

[8]  Ma Zeng-qiang,et al.  Short-Term Traffic Flow Prediction Based on ANFIS , 2009, 2009 International Conference on Communication Software and Networks.

[9]  V M F Mendes,et al.  Hybrid Wavelet-PSO-ANFIS Approach for Short-Term Wind Power Forecasting in Portugal , 2011, IEEE Transactions on Sustainable Energy.

[10]  Maria Grazia De Giorgi,et al.  Error analysis of short term wind power prediction models , 2011 .

[11]  Tie Qiu,et al.  Recurrent Broad Learning Systems for Time Series Prediction , 2020, IEEE Transactions on Cybernetics.

[12]  Dejan J. Sobajic,et al.  Learning and generalization characteristics of the random vector Functional-link net , 1994, Neurocomputing.

[13]  Gabriel Rilling,et al.  On empirical mode decomposition and its algorithms , 2003 .