A typical scenario generation method for renewable energy power systems based on time-sequenced operational simulations

Traditional clustering methods based on source-load information struggle to accurately describe the time-sequenced operational characteristics of renewable energy power systems. To address this， a typical scenario generation method for renewable energy power systems based on time-sequenced operational simulations is proposed， considering both the AC power flow distribution and startup-shutoff strategies for units. Firstly， a unit combination model that incorporates AC power flow constraints is constructed， and a two-stage solution strategy based on second-order cone relaxation is adopted to perform 8，760-hour time-sequenced simulations in the multi-period optimal power flow model. Secondly， based on data characteristics such as renewable energy output levels， line congestion conditions， and unit startup-shutoff from the time-sequenced simulation results， an improved K-means algorithm is employed to extract typical scenarios. Finally， the validity of the proposed method is verified through a case study， providing a reference for renewable energy power system planning and scheduling.