Power distribution network single-phase earth fault category probability prediction method

The invention relates to the technical field of power distribution network fault detection and protection, and discloses a power distribution network single-phase earth fault category probability prediction method, which comprises the steps of S1, constructing a plurality of fault features reflecting earth faults according to a mainstream method, and forming a fault feature set X; s2, in combination with a multi-layer perceptron mode, constructing a single-phase earth fault category identification learning model of a mapping relation between fault features and various fault categories, and introducing various activation functions into an output layer of the single-phase earth fault category identification learning model, probabilistic conversion is carried out on the result of the output layer obtained through the activation function; and S3, introducing a cross entropy mode to carry out reverse weight and deviation correction on the single-phase earth fault category identification learning model, so that the