Inherent Spatiotemporal Uncertainty of Renewable Power in China

Abstract

Abstract Solar and wind resources are vital for the sustainable and cleaner transition of the energy supply. Although renewable energy potentials are assessed in the literature, few studies examine the statistical characteristics of the inherent uncertainties of renewable generation arising from natural randomness, which is inevitable in stochastic-aware research and applications. Here we develop a rule-of-thumb statistical learning model for wind and solar power prediction and generate an hourly and year-long dataset of prediction errors in 30 provinces of China. The results reveal diversified spatial and temporal distribution patterns of prediction errors, indicating that more than 70% of wind prediction errors and 50% of solar prediction errors arise from scenarios with high utilization rates. We discover that the first-order difference and peak ratio of generation series are two primary indicators explaining the distribution characteristics of prediction errors. Furthermore, the prediction errors could result in additional CO2 emissions from coal-fired thermal plants. We estimate that such emission would potentially reach 319.7 megatons in 2030, accounting for 7.7% of China’s power sector. Finally, improvements in investment incentives and interprovincial scheduling could be suggested.