I. Introduction
In the last few decades, due to global warming and the energy crisis, wind power generation rapidly has expanded globally as a renewable energy source [1]. However, the typical randomness and volatility of wind energy are significant challenges to operation stability and the economic dispatch of modern power systems [2]. An effective solution to lessen the unfavorable influence of these attributes is to apply cutting-edge prediction approaches for wind power [3]. Given the above severe challenges, many engineers and researchers have carried out extensive research on wind power forecasting based on physical, statistical, machine learning, and hybrid models.