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

Power transmission line diagnosis method and system based on multi-task deep convolutional neural network

A technology of transmission lines and deep convolution, applied in biological neural network models, neural architectures, information technology support systems, etc., can solve problems such as complex model structure, poor generalization ability, and long online learning and training time, and achieve generalization Strong ability, fast training speed, simple and stable network structure

Active Publication Date: 2020-10-30
NARI TECH CO LTD +1
View PDF3 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] In order to solve the deficiencies in the prior art, the present invention provides a transmission line diagnosis method and system based on a multi-task deep convolutional neural network, which solves the problem of long online learning and training time and poor generalization ability of the existing transmission line diagnosis method. Problems with complex model structures

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 transmission line diagnosis method and system based on multi-task deep convolutional neural network
  • Power transmission line diagnosis method and system based on multi-task deep convolutional neural network
  • Power transmission line diagnosis method and system based on multi-task deep convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0047] A transmission line diagnosis method based on a multi-task deep convolutional neural network, comprising steps:

[0048] Step 1. After preprocessing the picture samples of abnormal objects in the transmission line corridor, expand the sample set. The sample set includes: training set and test set;

[0049] By performing various morphological operations on the image samples of abnormal objects in the transmission line corridor acquired by the existing video acquisition device, the samples used for training and learning are artificially increased. It enables the network model to learn sample features in different situations, greatly improving the robustness of the system;

[0050] Image sample preprocessing of abnormal objects in transmission line corridors, including:

[0051] 1) Extract the image corresponding to the rectangular frame of the abnormal object after screening the original image samples to obtain the original effective image sample; (the image sample of th...

Embodiment 2

[0081] A transmission line diagnosis system based on a multi-task deep convolutional neural network, including:

[0082] The difference image acquisition module is used to acquire the transmission line image to be diagnosed, and compare it with the normal image to obtain the difference image;

[0083] The recognition module is used to input the difference image into the pre-trained network model for category classification of abnormal objects and the network model for attribute classification of abnormal objects to identify the attributes and types of abnormal objects;

[0084] The network model for attribute classification of abnormal objects is obtained by retraining the remaining network layers based on the characteristic parameters of some convolutional layers in the network model for classification of abnormal object categories by MMD distance migration.

[0085] Further, the acquisition process of the network model for the attribute classification of abnormal objects is ...

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 transmission line diagnosis method and system based on a multi-task deep convolutional neural network, and the method comprises the steps: obtaining a to-be-diagnosed power transmission line image, comparing the to-be-diagnosed power transmission line image with a normal image, and obtaining a difference image; inputting the difference image into a pre-trained network model for abnormal object category classification and a network model for abnormal object attribute classification, and identifying attributes and types of abnormal objects, wherein the network model for abnormal object attribute classification is obtained by migrating part of convolution layer characteristic parameters in the network model for abnormal object category classification based on MMD distance and retraining the remaining network layers. The method is high in training speed, stable and simple in model and convenient to expand.

Description

technical field [0001] The invention relates to the technical field of power system operation and fault diagnosis, in particular to a transmission line diagnosis method and system based on a multi-task deep convolutional neural network. Background technique [0002] With the rapid development of the national economy, the demand for power resources in all walks of life is also increasing day by day. As the lifeline of the economy, the safe and stable operation of the power grid has an important impact on people's lives, industrial production and even the normal operation of society. As the link connecting power plants, substations and users, transmission lines are an important part of the power grid and the main artery of energy transmission in the whole society. The safety and reliability of transmission line operation have received more and more attention. Installing video surveillance equipment on the transmission line, realizing the prediction of the abnormal situation o...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/00G06N3/04G06K9/62
CPCG06T7/001G06T2207/20081G06T2207/20084G06N3/045G06F18/2413Y04S10/50
Inventor 张恒饶丹李临风周华良李友军尹宇轩王军张吉
Owner NARI TECH CO LTD
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