Data-driven hybrid approaches for renewable power prediction toward grid decarbonization: Applications, issues and suggestions

Abstract Global warming and climate change are serious problems that need urgent action and replacement. Renewable power could be the promising alternative solution to fossil fuel-based electricity generation in minimizing carbon intensity and achieving the global decarbonization target by 2050. However, intermittent characteristics of renewables such as solar and wind have resulted in negative effects on the operation, reliability, and stability of the power grid. To address these concerns, the hybridization of data-driven algorithms has achieved substantial contributions in renewable power prediction with regard to efficiency, precision and robustness. The main contribution of this study is to provide a detailed explanation of the recent progress of hybrid data-driven algorithms for renewable power prediction including solar, wind, ocean, hydro, and geothermal highlighting their variables, forecasting horizons, performance indexes, contributions and limitations. Besides, the impact of grid decarbonization in connection with renewable power is analyzed rigorously. Furthermore, this review explores the key issues and challenges of hybrid data-driven approaches in renewable power prediction to identify existing research gaps and limitations. Finally, this paper delivers selective suggestions that will support academic researchers and power engineers to develop advanced hybrid data-driven approaches for future renewable power prediction toward achieving the decarbonization goal.