•  
  •  
 

Keywords

power grid; rough set theory; Bayesian network; HHT; evidence theory

Abstract

The safety and stable operation of power grid is the prerequisite for reliable transmission, transformation, and distribution. Therefore, when the power grid fails, it is very important to locate the fault quickly and accurately and shorten the fault time. Firstly, the information of component switching value and electrical quantity is obtained from the relevant monitoring system of the power grid. The initial decision table of relevant switching value information is formed according to the fault area, and the effective signal of electrical quantity information is extracted. Then, the rough set theory, Bayesian network, Hilbert-Huang transform (HHT), and other theories are used to calculate the component fault degree and distortion degree. Subsequently, the improved D-S evidence theory is employed to fuse the fault degree of component switching value with the distortion degree of electrical quantity. Finally, the local topology of a regional power grid is used to test the improved Bayesian network model. The simulation results show that the model can improve the diagnosis speed. The IEEE 39 node is used as an example, and it is verified that the introduction of switching value can improve the diagnostic accuracy, and data fusion reduces the uncertainty in the evaluation model.

DOI

10.19781/j.issn.1673-9140.2025.02.005

First Page

42

Last Page

49

Download

Share

COinS