Distributed Photovoltaic Power Cluster Partition Based on DPC Clustering Algorithm

With the distributed access of large-scale renewable energy, the distribution network is facing more and more serious problems such as overvoltage and power backfeeding. This puts higher demands on the rapidity of voltage control in the current tidal state. Considering the complex control process of voltage centralized control method and the long time consuming traditional zoning method. This paper first analyses the impact of active and reactive power on the voltage of the distribution network after distributed PV access and uses the combined sensitivity as an electrical distance to determine the closeness of the electrical connection between nodes. Secondly, according to the calculated electrical distance as the input of DPC clustering algorithm, the distribution network nodes are clustered to achieve fast partitioning of the distribution network. Finally, the concept of complex network modularity function is introduced to evaluate the partitioning results to obtain the optimal partitioning results. The effectiveness of the method is demonstrated by simulation experiments using IEEE 33 nodes.