An Advanced Peaking Method for Improved Hydropower Plant Regulation and Power Grid Management

Abstract

Hydropower, as a crucial component of power grid systems, plays an essential role in peak regulation due to its fast start-stop and high-speed climbing capabilities. Current hydropower peak regulation methods struggle to consider complex load demand and the highly coupled characteristics of runoff simultaneously. This study proposes the Adaptive Segmented Cutting Load Algorithm (ASCLA) to restructure the power station's load process and segment the scheduling period based on load characteristics, ensuring hydropower stations operate in peak regulation mode throughout the entire cycle. The method determines each sub-scheduling period's peak regulation depth based on runoff characteristics and considers factors impacting peak regulation capability. To minimize the residual load's rolling data window standard deviation, we apply ASCLA to the Three Gorges Reservoir (TGR) simulation. We introduce four evaluation indicators: Mean Squared Deviation of the Rolling Window (MSDRW), total time variation of residual load, peak residual load, and response time to assess peak regulation effectiveness. Our method can handle peak regulation demands under varying runoff conditions, providing feasible scheduling solutions. Simulations and analyses reveal ASCLA demonstrates stronger load tracking ability, a broader adjustment range of load peaks and valleys, and a more significant peak regulation effect compared to the conventional method. Finer segmentation of sub-scheduling periods and final water level determination under conditions of higher load variability and drier runoff optimizes the power station's regulation capacity and meets the power grid's operational needs. In conclusion, our research develops a comprehensive and adaptable peak regulation scheduling model for hydropower stations, offering more effective solutions to address challenges related to extreme weather events and renewable energy integration.