Journal
ELECTRIC POWER SYSTEMS RESEARCH
Volume 211, Issue -, Pages -Publisher
ELSEVIER SCIENCE SA
DOI: 10.1016/j.epsr.2022.108186
Keywords
Point prediction; Decomposition-selection-ensemble prediction; system; Interval forecasting; Sub-model selection strategy; Wind speed forecasting
Categories
Funding
- Major Program of National Fund of Philosophy and Social Science of China [19ZDA120]
Ask authors/readers for more resources
In this study, a novel wind speed prediction system is proposed, which can conduct point and interval prediction simultaneously. The system successfully integrates the merits of component models and effectively overcomes the disadvantages of traditional prediction methods. Simulation results demonstrate its important application value in the scheduling and management of power systems.
Wind energy is a clean, efficient, and eco-friendly energy with promising application prospects. In the wind power system, reliable wind speed forecasting is crucial as it will better integrate wind energy into the power system and boost the safe operation of power grid. Scholars have developed several wind speed prediction methods; however, these methods usually ignore the importance of sub-model selection and interval prediction. In this study, a novel decomposition-selection-ensemble prediction system, which comprises a decomposition strategy, sub-model selection, system optimization, prediction, and assessment, is proposed to simultaneously conduct point and interval prediction. The proposed system successfully integrates the merits of component models and effectively overcomes the disadvantages of traditional prediction methods. The simulation result reveals that the proposed system can assign an optimal model for each sub-series and significantly promote an improvement of wind speed forecasting ability, demonstrating its superior application value in the scheduling and management of power systems.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available