Portability of Short Term Wind Power Forecasting

Wind power forecasting (WPF) plays an increasingly important role in integrating wind power into electricity systems. The porta-bility of state-of-the-art WPF methods is assessed in this paper; the goal is to explore the influence of model hyperparameter configuration on forecasting performance. The overall aim is to expand understanding of WPF methods in order to improve the reliability and competitiveness of wind energy generation technologies. The performance of two hybrid decomposition-based WPF methods are evaluated and compared, Variational Mode Decomposition & Feed Forward Neural Network (VMD-FFNN) and Ensemble Empirical Mode Decomposition & Feed Forward Neural Network (EEMD-FFNN). Supervisory Control and Data Acquisition (SCADA) data from wind farms in Ireland and the UK is utilised. The robustness and portability of the WPF methods when applied to different datasets is examined. Both WPF methods produce robust and accurate forecasts across different datasets however, forecasts produced using low-resolution data are superior to high-resolution data forecasts. In the portability assessment, the forecasting performance is sensitive to two of the four model hyperparameters examined.