A Construction Method for Typical Source-Load Chronological Scenarios Integrating Multiple Critical Characteristic Dimensions
Wei Lou, Xiaohu Zhang, Xu Ling, Mengyao Zhang, Heng Zhang, Haozhong Cheng, Haipeng Kang · 2025
The construction of typical source-load chronological scenarios is crucial for power system optimization scheduling and planning. Existing research on typical scenario construction involves limited dimensions and insufficient rationality verification indicators, which may lead to higher operational or planning costs for power systems. Thus, this study proposes a construction method for typical source-load chronological scenarios integrating multiple critical characteristic dimensions. First, considering the spatiotemporal correlations among wind, solar, and load, a scenario generation model based on multivariate joint probability distribution is proposed to generate sufficient scenario samples. Then, a multidimensional critical characteristic dimension identification method based on correlation analysis is proposed to extract indicators characterizing the critical features of wind-solar-load. Next, a typical scenario construction process based on density clustering algorithm is presented, using the critical characteristic indexes as clustering feature dimensions. Finally, scenario validity verification indicators are proposed. The proposed method is validated using wind-solar-load data from an actual system.