Reactive Power Optimization of Distribution Network Based on BP Neural Network and Intelligent Scene Matching

Reactive power optimization plays an important role in ensuring the safety and economic operation of distribution network. This paper proposes a reactive power optimization method based on BP neural network and intelligent scene matching, which does not depend on the model and parameters of distribution network, greatly improves the decision-making efficiency. The method classifies scenes according to three optimization objectives, uses historical scene library generated based on big data during the distribution network operation as a training set and the scenes at the moment to be optimized as a test set, match the existing scenes in the historical scenes based on BP neural network technology, and the corresponding control strategies are taken as the reactive power optimization strategy at the moment to be optimized, so as to reduce active power loss and voltage offset. Finally, the modified IEEE 30-bus system was connected to random loads such as photovoltaics, energy storage devices and electric vehicle charging stations to verify the effectiveness and correctness of the models and methods.