Monthly Wind Power Forecasting: Integrated Model Based on Grey Model and Machine Learning

Wind power generation has been developed rapidly due to rising global interest in renewable clean energy sources. Accurate prediction of the potential amount such is great significance development. As wind changes greatly by season, time series analysis considered as a natural approach characterize seasonal fluctuation and exponential growth. In this paper, dual integrated hybrid model presented using random forest (RF) incorporate extreme gradient boosting (XGB) with empirical mode decomposition (EMD) fractional order accumulation grey (FSGM). For vertical dimension processing, decomposed into high low frequency components. Then, components are predicted XGB learning machine (ELM), respectively. growth horizontal FSGM applied same month different years. Consequently, proposed can not only be used capture trend but also investigate complex high-frequency variation. To validate model, it analyze characteristics for China from 2010 2020, results compared popularly known models; illustrate that superior other models examining series.