Coordinated planning method for large-scale charging piles and reactive power compensation based on deep reinforcement learning

Abstract

The application provides a large-scale charging pile and reactive power compensation collaborative planning method based on deep reinforcement learning, which comprises the following steps: S1. determining the topological structure in the power grid in the target area, and dividing the target area into a plurality of sub-areas according to the set distance; S2. constructing an active power and reactive power calculation model of the current node after considering the newly added load of the charging pile and the reactive power compensation of the reactive power generator; S3. constructing an alternating current flow equation based on the active power and reactive power calculation model of the current node, and determining the voltage of the current node; S4. constructing a planning model of the charging pile and the reactive power generator; S5. constructing a deep reinforcement learning neural network, inputting the planning model of the charging pile and the reactive power generator and the voltage of the current node into the deep reinforcement learning neural network, and predicting the number of newly added charging piles and the configuration capacity of the newly added reactive power generator in the target area by the deep reinforcement learning neural network.

Images