Dynamic Grouping Equivalent Modeling of Regional Distributed Photovoltaic Power Generation System Based on Improved FCM Algorithm

To model large-scale regional dispersed photovoltaic systems and study their operational characteristics, a dynamic clustering equivalent modeling method for regional dispersed photovoltaic power generation systems is proposed. Firstly, feature points are selected as clustering indicators from the fault curves that can reflect the dynamic characteristics of the photovoltaic system. Then, an improved FCM clustering algorithm is used to cluster the regional decentralized photovoltaic power generation system and construct a multi machine equivalent model for the regional decentralized photovoltaic system. Among them, in order to improve the traditional fuzzy C-means clustering algorithm (FCM), which is prone to falling into local optima due to excessive dependence on the initial clustering center point, the FCM algorithm is preselected with the initial clustering center before use, and then clustering is carried out. Finally, the rationality and accuracy of the proposed equivalent modeling method were verified through the Simulink simulation platform.