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

Transmission line equipment defect detection method based on sample offset network

A transmission line and defect detection technology, applied in the field of image data processing and neural network, can solve problems such as being easily affected by personal experience, high professional requirements, and high risk, so as to reduce over-fitting, enhance detection ability, and facilitate The effect of the operation

Active Publication Date: 2022-08-09
NARI INFORMATION & COMM TECH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The initial fault inspection method mainly relied on on-site staff to conduct manual analysis and diagnosis based on manual inspection experience. This method has a high level of professional requirements for the staff, and is often accompanied by shutdown inspections, which consumes a lot of manpower, material and financial resources, takes a long time, and is dangerous. Highly sexual and easily influenced by personal experience

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
  • Transmission line equipment defect detection method based on sample offset network
  • Transmission line equipment defect detection method based on sample offset network
  • Transmission line equipment defect detection method based on sample offset network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0062] The invention obtains a transmission line equipment defect detection model based on a preset training set and deep learning structure training, and can detect six defect types of bird's nest, insulator self-explosion, anti-vibration hammer damage, tower base buried, tower base flooding, and bird shield damage. detection.

[0063] figure 1 The above is the overall architecture diagram of defect detection. The collected image data is subjected to image preprocessing and input to the convolutional neural network model for model training. The network is divided into three parts: feature extraction, candidate frame extraction and positioning and classification. After many iterations , the network parameters are optimized, and finally the trained convolutional neural network model is obtained. The test set images are input into the trained convolutional neural network model for testing, and the overall performance of the convolutional neural network model is judged by detect...

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 transmission line equipment defect detection method based on a sample offset network. The data set is sent into a designed convolutional neural network model for training, and the self-parameters of the convolutional neural network model are improved; The convolutional neural network model is deployed on the detection equipment, and the transmission line equipment defects are detected. The invention uses the mosaic data enhancement method to process data, which can supplement the missing images of the samples, enrich the background information of defect categories, and reduce the overfitting of the network. Through the feature extraction and feature fusion modules, the diversity of input images is enriched, so that the network can accurately determine the region of interest and enhance the detection ability of the network. By correcting the position of the candidate frame, the classification task and the regression task can obtain different candidate regions. The number of actions recognized by the recognition method is scalable, and the extension operation is simple and easy for developers to operate.

Description

technical field [0001] The invention relates to a defect detection method for transmission line equipment based on a sample offset network, belonging to the technical field of image data processing and neural network in the field of artificial intelligence. Background technique [0002] Due to the increasing scale of my country's power grid, the quantity and quality of on-site inspections of electric power have higher requirements for electric power personnel. However, the transmission and electric power site environment is complex, with a variety of equipment distributed, and it is difficult to accurately observe the connection between equipment through various overhead lines. , to ensure the stable operation of the transmission line, all equipment needs to work together properly. The initial fault inspection method mainly relies on manual analysis and diagnosis by on-site staff with manual inspection experience. This method requires a high level of professionalism for the s...

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 Patents(China)
IPC IPC(8): G06V10/25G06V10/774G06V10/80G06V10/82G06K9/62G06N3/04G06Q50/06G07C1/20
CPCG06Q50/06G07C1/20G06N3/045G06F18/253G06F18/214Y04S10/50
Inventor 毛进伟罗旺陈海鹏
Owner NARI INFORMATION & COMM TECH
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