Patent · US Active

Methods, systems, and computer readable mediums for determining a system state of a power system using a convolutional neural network

US11544522B2 · kind B2 · utility

0Cited by
0References
14Claims
0Family size

Assignee

Inventors

Key dates

Filing dateApr 11, 2019
Grant dateJan 3, 2023
Priority date
Expiry dateNov 3, 2041

Classification

  • Technology area (CPC Y)Emerging Cross-Sectional Technologies
  • CPC primaryY04S40/20
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

Methods, systems, and computer readable mediums determining a system state of a power system using a convolutional neural network using a convolutional neural network are disclosed. One method includes converting power grid topology data corresponding to a power system into a power system matrix representation input and applying the power system matrix representation input to a plurality of convolutional layers of a deep convolutional neural network (CNN) structure in a sequential manner to generate one or more feature maps. The method further includes applying the one or more feature maps to a fully connected layer (FCL) operation for generating a respective one or more voltage vectors representing a system state of the power system.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.