Power network node vulnerability assessment method based on BP neural network

Abstract

The application discloses a power network node vulnerability evaluation method based on a BP neural network, and comprises the following steps: building a power training network and a power test network model; acquiring an electrical characteristic matrix; constructing a characteristic data extraction algorithm; acquiring a vector dimension unified electrical characteristic matrix; acquiring an electrical abstract characteristic matrix of a power network node; defining a node electrical characteristic label; constructing a node characteristic attribute; constructing a to-be-normalized data set and normalizing; building a BP neural network training model; inputting the power test network into the BP neural network training model to obtain a test value after inverse normalization processing; and acquiring a priority protection node set to apply key protection; the method comprehensively considers a power network topological structure and electrical characteristics, extracts electrical characteristic dimensions uniformly, is suitable for high-dimensional, nonlinear and large-scale complex power grid vulnerability evaluation, is universal for all power grids, quickly and accurately obtains a priority protection set, and can effectively improve the safe and stable operation efficiency of a power system.

Images