Distributed power supply

electric vehicles and energy storage are connected to the distribution network at a high proportion. This improves the intelligence and controllability of the distribution network

and also brings forward more complex economic and security requirements for the optimization and regulation of the network. There is a problem of online optimal control of the distribution network without real-time measurement

so an optimal control scheme for the network based on feature matching and sensitivity-assisted decision making is proposed. First

a distribution network historical feature database and strategy database considering different operational characteristics are constructed to improve the accuracy and speed of feature matching. Network loss and voltage fluctuation are effectively reduced through source-network-load-storage coordination optimization. Secondly

the generation method of the network matching strategy based on feature matching is proposed. This removes the limitations of the power flow model and greatly improves the efficiency of real-time optimization. Finally

in order to correct the strategic error caused by the feature matching deviation

an auxiliary decision-making method for online optimization control of the network is proposed. A command weight coefficient based on the system matching deviation rate is designed to improve the accuracy of online control instructions. The accuracy and feasibility of the proposed scheme are verified by simulation examples.