Data dimension reduction method and system for distribution network self-healing test model, terminal and medium

Abstract

The invention discloses a data dimension reduction method and system for a distribution network self-healing test model, a terminal and a medium, and the method comprises the steps: obtaining an input data set and output electrical quantity data in a self-healing test process, and constructing an input matrix and an electrical quantity output matrix of the self-healing test model; normalizing the input matrix and the electrical quantity output matrix; calculating a first correlation coefficient and a second correlation coefficient of the normalized input matrix and the electrical quantity output matrix; determining a comprehensive correlation coefficient according to the first correlation coefficient and the second correlation coefficient; and setting a threshold value for the comprehensive correlation coefficient to carry out data selection of an input matrix so as to realize data dimension reduction. The correlation coefficient of the input matrix and the output electrical quantity of the self-healing test set is accurately calculated from the perspective of variation trend and distance correlation, a more accurate dimension reduction result is achieved compared with an existing self-healing test model, and higher recognition accuracy is achieved compared with a single data dimension reduction method.

Description

technical field [0001] The invention relates to the technical field of distribution automation, in particular to a data dimensionality reduction method, system, terminal and medium of a distribution network self-healing test model. Background technique [0002] Due to the low voltage level and many stages of the distribution network, the transmission power and distance of a single feeder are generally not large, the mixed network of cable lines and overhead lines brings complexity to the operation and analysis of the power grid, and the operation mode in the distribution network is changeable And other characteristics, leading to the typical characteristics of the distribution network such as large network scale, complex structure, and highly asymmetric structure and parameters. With the large-scale application of microgrid and distributed power, the stability of distribution network is further affected, and self-healing control technology is facing more severe challenges. ...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a data dimensionality reduction method, system, terminal and medium of a distribution network self-healing test model, so as to solve the existing self-healing control technology with too high input data dimension, slow self-healing test process, and accurate test sexual inadequacy

Images