Abstract:With the large-scale access of distributed resources such as photovoltaics, electric vehicles, and energy storage to rural power distribution systems, it brings great challenges to the operation and regulation of system. Predicting and evaluating the active and reactive power support capacity of rural distribution systems with high proportion of distributed resources has become an important issue to ensure the safe and stable operation of the power grid. Therefore, this paper proposes an Informer prediction model based on Transformer, which uses the maximum mutual information and historical power data to accurately predict the power data of future points. The voltage support degree and reactive power regulation ability of the node are evaluated by using the sensitivity of the node voltage and the adjustable reactive power capacity. At the same time, new indicators and dynamic indicators are introduced to quantify and evaluate the active and reactive power support capabilities of the distribution system. Finally, the simulation experiment is carried out in a rural power distribution system with high proportion of distributed resources in Guizhou. The results show that the optimal control method proposed in this paper has better safety and economy, which is conducive to promoting the construction and operation of the new rural power distribution system.