Ultra-Short-Term Operating Reserve Requirement Assessment of Power System Based on Improved XGboost Quantile Regression

As the installed capacity of renewable energy sources explosively increases, the current deterministic reserve standards are no longer suitable for the high proportion of renewable energy integration and the need for safe and stable operation of the power grid. It is urgent to improve the practical level of ultra-short-term operating reserves. This article proposes an assessment method for the ultra-short-term operating reserve requirements of the power system based on QRXGboost-RSA, which combines the XGboost model with quantile theory and adopts RSA to optimize the model, assessing the ultra-short-term operating reserve requirements for future periods at different quantile points. Finally, simulated verification is conducted on a dataset from a province in Northwest China, and the results indicate that the proposed method can effectively assess ultra-short-term operating reserve requirement.