Power grid dispatching drawing automatic identification method and system based on improved YOLOv3 network and medium

A power grid dispatching and automatic identification technology, applied in the direction of neural learning methods, character and pattern recognition, biological neural network models, etc., can solve the problems of large base of power grid drawings and data, and achieve the solution of mean value shift, good training effect, increase the sticky effect

Inactive Publication Date: 2020-01-07
STATE GRID HUNAN ELECTRIC POWER +1
View PDF2 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, in view of the large base and rapid growth of power grid drawings, the existing raster scanning technology cannot fundamentally solve the problem of management expansion of drawings and data. How to realize accurate automatic identification has become a key technical problem to be solved urgently

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 grid dispatching drawing automatic identification method and system based on improved YOLOv3 network and medium
  • Power grid dispatching drawing automatic identification method and system based on improved YOLOv3 network and medium
  • Power grid dispatching drawing automatic identification method and system based on improved YOLOv3 network and medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] Such as figure 1 In this embodiment, the implementation steps of the automatic identification method for power grid dispatching drawings based on the improved YOLOv3 network include:

[0031] 1) Input the image of the grid dispatching drawing, and preprocess the image of the grid dispatching drawing;

[0032] 2) Input the preprocessed power grid dispatching drawing image into the improved YOLOv3 network that has completed the training, and obtain the device recognition result corresponding to the power grid dispatching drawing image. The improved YOLOv3 network is trained to establish the relationship between the power grid dispatching drawing image and the corresponding device recognition result. Mapping relationship; the activation function of the improved YOLOv3 network uses the improved leaky ReLU activation function, and the linear function of the improved leaky ReLU activation function in the negative interval is replaced by the tanh function, such as figure 2 s...

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 grid dispatching drawing automatic identification method and system based on an improved YOLOv3 network and a medium. The method comprises the following steps: an inputpower grid dispatching drawing image is preprocessed, and the preprocessed power grid dispatching drawing image is input into a trained improved YOLOv3 network to obtain an equipment identification result corresponding to the power grid dispatching drawing image; an improved YOLOv3 network is trained to establish a mapping relationship between a power grid dispatching drawing image and a corresponding equipment identification result; an improved leeky ReLU activation function is adopted as the activation function; a linear function in a negative interval is replaced by a tanh function; the characteristics of an original activation function are reserved, mean shift caused by a negative interval is solved, and in addition, the tanh function of the negative interval is a non-linear function,so that the method is not excessively sensitive to special data and wrong abnormal data during network training, the robustness is improved, and the method has the characteristic of soft saturation.

Description

technical field [0001] The invention relates to an automatic recognition technology for power grid dispatching drawings, in particular to an automatic recognition method, system and medium for power grid dispatching drawings based on an improved YOLOv3 network. Background technique [0002] With the continuous development of algorithm computing power and the gradual maturity of data systems, State Grid is also actively applying artificial intelligence to power grids, and is also constantly making technological innovations. With the gradual intelligentization of the power grid and the new round of rural power grid transformation and upgrading, it is constantly found that there are a large number of graphic files in grid dispatching, and the growth rate is extremely fast. These important and complicated original drawings are both the cornerstone and the foundation for the construction of smart grid. stumbling block. Therefore, how to convert the original drawing data into gra...

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): G06K9/00G06N3/04G06N3/08
CPCG06N3/08G06V30/422G06N3/048
Inventor 袁文谢培元刘永刚孟朝晖潘飞来唐云红杨俊刘力姜新凡曾次玲吕风仪龙立波张拯姜学皎王蕾
Owner STATE GRID HUNAN ELECTRIC POWER
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap