Fig. 3: Peaks distribution and the impact on the wind and solar power prediction errors.
From: Inherent spatiotemporal uncertainty of renewable power in China
a Influence of the wind hourly peaks. The radius of each bubble indicates the ratio of the wind daily peaks. b Wind hourly peak distribution in 10 power generation intervals for Tianjin (TJ), Shandong (SD), Shanxi (SX), and Gansu (GS). c Influence of the solar daily peaks. The radius of each bubble represents the ratio of the solar hourly peaks. d Solar daily peak distribution in 10 power generation intervals for Beijing (BJ), Jiangsu (JS), Hubei (HB), and Inner Mongolia (IM). In (a) and (c), the number of bubbles is 30, representing the 30 provinces of China, excluding Tibet (wind), Chongqing (solar), Hong Kong, Macao, and Taiwan. The black linear regression line fits the center of the bubbles, complemented by the slope, intercept, and coefficient of determination (CoD). The color of each bubble indicates the different categories: red—category with the largest prediction error; yellow—category with the second-largest prediction error; blue—category with the third-largest prediction error; green—category with the smallest prediction error.
