Multi-Horizon Data-Driven Wind Power Forecast: From Nowcast to 2 Days-Ahead

Generation uncertainty is an obvious challenge posed by renewable energy sources such as wind power with effects spawning from stability threats, to economic losses. Data-driven forecasting methods draw increasing attention due to the amount of data available, flexibility and cost-effectiveness among other factors. However, there are concerns regarding effective feature selection and tuning of these models since common naive approaches focus on Pearson or Shapley. This papers uses the development of an active power forecaster for a wind turbine to conduct a thorough sensitivity analysis addressing how different sampling rates, machine learning (ML) methods, features and hyperparameters influence accuracy. Which is computed with the Root-Mean Squared Error and compared against Persistence. The selected ML-methods are Random Forest and Long-Short Term Memory Artificial Neural Networks. The forecasters are multi-horizon & multi-output model targeting 1 minute, 1 hour, 5 hours and 2 days ahead by using sampling rates of 1 second, 1 minute, 5 minutes and 1 hour respectively. The results show which method is more suitable for which horizon and provides insight into which features reduce RMSE of the best performers, whose average is 10, 13, 17 and 25 % for each horizon respectively. The conclusions of the sensitivity analysis can be applied for regions with highly volatile weather, such as coastal areas.

[1]  Elisa Mussumeci,et al.  Large-scale multivariate forecasting models for Dengue - LSTM versus random forest regression. , 2020, Spatial and spatio-temporal epidemiology.

[2]  Jin Hur,et al.  An Ensemble Forecasting Model of Wind Power Outputs Based on Improved Statistical Approaches , 2020, Energies.

[3]  Jianchun Peng,et al.  A review of deep learning for renewable energy forecasting , 2019, Energy Conversion and Management.

[4]  Platon Patlakas,et al.  Development of a High-Resolution Wind Forecast System Based on the WRF Model and a Hybrid Kalman-Bayesian Filter , 2019, Energies.

[5]  Javier Béjar,et al.  “Dust in the Wind...”, Deep Learning Application to Wind Energy Time Series Forecasting , 2019, Energies.

[6]  Farhan Ullah,et al.  A New Hybrid Approach to Forecast Wind Power for Large Scale Wind Turbine Data Using Deep Learning with TensorFlow Framework and Principal Component Analysis , 2019, Energies.

[7]  Anton Kaifel,et al.  Minute-Scale Forecasting of Wind Power—Results from the Collaborative Workshop of IEA Wind Task 32 and 36 , 2019, Energies.

[8]  Igor Melnyk,et al.  Deep learning algorithm for data-driven simulation of noisy dynamical system , 2018, J. Comput. Phys..

[9]  Dabao Zhang,et al.  A Coefficient of Determination for Generalized Linear Models , 2017 .

[10]  Monidipa Das,et al.  Data-driven approaches for meteorological time series prediction: A comparative study of the state-of-the-art computational intelligence techniques , 2017, Pattern Recognit. Lett..

[11]  Chongqing Kang,et al.  A Short-Term Wind Power Forecasting Approach With Adjustment of Numerical Weather Prediction Input by Data Mining , 2015, IEEE Transactions on Sustainable Energy.

[12]  Felix F. Wu,et al.  Uncertainty management in power system operation , 2015 .

[13]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[14]  Marion P. Mittermaier,et al.  The Potential Impact of Using Persistence as a Reference Forecast on Perceived Forecast Skill , 2008 .

[15]  Tarlochan Kaur,et al.  Application of artificial neural network for short term wind speed forecasting , 2016, 2016 Biennial International Conference on Power and Energy Systems: Towards Sustainable Energy (PESTSE).

[16]  Tao Li,et al.  Review of Evaluation Criteria and Main Methods of Wind Power Forecasting , 2011 .