Source-Load-Storage Integration Optimization Based on Incremental Load Multi-Dimensional Portrait

With the extensive integration of user-side resources, such as distributed renewable energy, electric vehicle charging stations, and energy storage systems, the volatility and randomness of the regional distribution networks have been intensified. The interaction characteristics between incremental loads and power grid are increasingly difficult to accurately characterize, posing significant challenges to optimizing the coordinated integration of source, load, and storage across different planning scenarios. Therefore, this study proposes a source-load-storage integration optimization method based on incremental load multi-dimensional stereoscopic profiling approach. Firstly, the Variational Bayesian Gaussian Mixture Model (VBGMM) clustering algorithm is employed to construct the interactive external characteristic index, and an incremental load multi-dimensional stereo portrait is built based on the correlation-based feature selection principle and a best-first search strategy. Finally, according to the analysis results of the incremental load profile, the load regulation potential and response capability are determined as the fundamental input data of the model optimization, and the bi-level programming model for source-load-storage integration is established. The normalized normal constraint method (NNC) is used to solve the problem. The simulation results demonstrate that the proposed algorithm can improve the carrying capacity and economic benefits of the regional distribution network, promote peak load shifting, and maintain voltage stability.