Extraction of Representative Scenarios for Photovoltaic Power with Shared Weight Graph Clustering
Xueqian Fu,Na Lu,Hongbin Sun,Youmin Zhang
DOI: https://doi.org/10.1109/tsg.2024.3411306
IF: 10.275
2024-01-01
IEEE Transactions on Smart Grid
Abstract:With the growing integration of photovoltaic (PV) generation, the operational conditions of power systems become more complex and variable. These intricate scenarios place significant pressure on the optimization calculations for power systems, necessitating the extraction of representative scenarios for PV power generation to enhance optimization efficiency. To address this issue, we have proposed a novel clustering model that extracts representative PV output scenarios through the fusion of adaptive feature weights and adjacent density weights. We propose an alternating optimization solution algorithm based on the Lagrange multiplier method and eigenvalue decomposition. The highlight of this work is the dual verification through theoretical proof and simulation experiment. In terms of theoretical proof, we analyze the sensitivity of clustering model parameters, demonstrate algorithm complexity, and theoretically prove the convergence of the proposed solution algorithm. Using actual PV output data from Australia, we validate the high cohesion, low coupling, noise resistance, and parameter sensitivity of the proposed clustering model, as well as the convergence of the proposed solution algorithm. The effectiveness of the proposed method in extracting representative scenarios of PV output has been confirmed through probabilistic power flow analysis using two IEEE test cases.