Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Power distribution network measurement data missing reconstruction method based on generative adversarial and double semantic perception

A technology for measuring data and distribution networks, applied to electrical components, circuit devices, AC network circuits, etc., can solve problems such as lack of reconstruction, insufficient local attention, difficulty in completely extracting power data features, and low data reconstruction accuracy , achieving good anti-noise performance and generalization ability, improving stability, and high data feature complexity

Active Publication Date: 2019-09-06
NORTH CHINA ELECTRIC POWER UNIV (BAODING)
View PDF3 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The purpose of the present invention is to provide a new generation-based Reconstruction method for missing measurement data of distribution network based on adversarial and dual semantic awareness

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Power distribution network measurement data missing reconstruction method based on generative adversarial and double semantic perception
  • Power distribution network measurement data missing reconstruction method based on generative adversarial and double semantic perception
  • Power distribution network measurement data missing reconstruction method based on generative adversarial and double semantic perception

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0041] like figure 1 As shown, the network structure of distribution network measurement data deletion reconstruction based on generative confrontation and dual semantic perception in this embodiment uses GAN to construct the main model, and builds a missing reconstruction network with two deep learning networks of generator and discriminator. build model.

[0042] like figure 2 As shown, in this embodiment, the generator and discriminator of game confrontation are constructed based on the idea of ​​generative confrontation, and feature extraction is performed on measurement data, the setting of layer parameters and the type of internal neural network are determined, and feature extraction is performed through a separate and iterative training method for the first time. Model feature extraction training, using the ADAM optimizer to optimize the hyperparameters of the generator and discriminator during the training process.

[0043] like image 3 As shown, this embodiment b...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a power distribution network measurement data missing reconstruction method based on generative adversarial and double semantic perception, and belongs to the technical field of electric power big data processing. According to the method, a novel double semantic perception missing reconstruction constraint is formed by combination of the contextual information, the generateperceptual information and the reconstruct perceptual information by means of actively extracting data characteristics and the spatio-temporal distribution characteristics by a model based on the concept of game countermeasure of a generative adversarial network, and the data with highest similarity to the to-be-reconstructed data is found to realize the measurement data missing reconstruction. The method disclosed by the invention is completely based on data driving, and overcomes the limitation in multiple aspects of adaptively, algorithm efficiency and accuracy in establishing a mathematical reconstruction model based on a service mechanism; by adoption of the method, distribution hypothesis and characteristic explicit modeling of data can be carried out by needing a large amount of prior knowledge in a traditional missing reconstruction process are avoided; and the defects of low convergence speed, unstable training and the like of the traditional GAN are overcome, and the methodhas relatively high data missing reconstruction precision.

Description

technical field [0001] The invention relates to a reconstruction method for lack of distribution network measurement data based on generative confrontation and dual semantic perception, and belongs to the technical field of electric power big data processing. Background technique [0002] In the context of energy reform and social development, distribution network, as an important infrastructure for economic and social development, plays a key role in realizing the strategic goals of smart grid and energy Internet. With the improvement of informatization, automation and interaction level of smart distribution network and the mutual penetration and integration with the Internet of Things, a large amount of data has been accumulated in the power system measurement system. Authentic and reliable data can correctly reflect the operating characteristics and objective laws of the power system. However, failures or interference may occur in all aspects of data collection, measureme...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): H02J3/00
CPCH02J3/00H02J2203/20
Inventor 齐林海杨玉莲王红
Owner NORTH CHINA ELECTRIC POWER UNIV (BAODING)
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com