Assembly tightening process sample collection system, deep learning network and monitoring system

A technology of deep learning network and sample collection, which is applied in the field of deep learning network and monitoring system, and sample collection system in the assembly and tightening process, which can solve the problem of not being able to notify the on-site assembly workers of abnormal situations at the first time

Active Publication Date: 2019-09-06
QINGDAO TECHNOLOGICAL UNIVERSITY
View PDF13 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the traditional monitoring method based on computer vision cannot make specific and accurate judgments on the forces and moments in the assembly process through the collected image information, and cannot promptly inform the on-site assembly workers of abnormal conditions in a timely manner

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
  • Assembly tightening process sample collection system, deep learning network and monitoring system
  • Assembly tightening process sample collection system, deep learning network and monitoring system
  • Assembly tightening process sample collection system, deep learning network and monitoring system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] see figure 1, assembling a tightening process sample collection system, including a video collection device 1, a wearable myoelectric device 2, a torque collection device 3 and a first computer 4, and the video collection device 1 collects video information and sends it to the first computer 4, so The myoelectric device 2 collects the myoelectric signals, inertial acceleration signals and orientation signals of the human body, and then sends them to the first computer 4, and the torque collection device 3 sends the collected torque information when the operator tightens the nut to the The first computer 4, the first computer 4 performs the following steps: identify the skeletal nodes of the human body in the video information, and calculate the coordinate information of each skeletal node; , orientation signal, torque information, and video information are all stored according to time tags; pictures are extracted from video information by time tags, and the pictures, my...

Embodiment 2

[0049] see Figure 4 , a deep learning network, using the sample library obtained by the sample collection system of embodiment 1, and training through the following training steps: using part of the sample data in the sample library as a training set, and dividing the sample data in the training set into multiple groups; The deep learning network includes an input layer, an output layer, a convolutional layer, and a fully connected layer. First, the input layer receives a set of integrated information sets and image information sets, and the sample data in the set of integrated information sets are subjected to the first volume The convolution operation of the product layer is expanded through the first fully connected layer to obtain the first one-dimensional feature value. The sample data of the set of image information is subjected to the convolution operation of the second convolution layer, and then passed through the second fully connected layer. After the connection la...

Embodiment 3

[0054] see Figure 6 and Figure 7 , a monitoring system for the assembly and tightening process, including an image acquisition device 6, an electromyography wearable device 7 and a second computer 8, the image acquisition device 6 collects and monitors the operation video of the on-site operator when assembling and tightening and converts it according to the set number of frames It is an image, and then sent to the second computer 8, the myoelectric wearable device 7 collects the myoelectric signal, inertial acceleration signal and orientation signal of the human body, and then sends it to the second computer 8, and the second computer 8 loads the embodiment In the deep learning network in the second, the second computer 8 performs the following steps: identify the skeletal nodes of the human body in the image, and calculate the coordinate information of each skeletal node; and orientation signals are input into the deep learning network, and the deep learning network outpu...

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 relates to assembly tightening process sample collection, which collects video information through a video collection device, collects an electromyographic signal, an inertial acceleration signal and an azimuth signal of a human body through an electromyographic device, collects torque information collected by a torque collection device, receives the torque information by a computerand stores the torque information according to a time label so as to obtain a sample library for machine learning. The invention further relates to a monitoring system for the assembly tightening process, and the second computer identifies human skeleton nodes in the image and calculates coordinate information of each skeleton node; the coordinate information of the bone nodes, the electromyographic signals, the inertial acceleration signals and the azimuth signals are input into a deep learning network; the monitoring torque value is output, the monitoring torque value is compared with a preset torque reference value, whether the tightening process is abnormal or not is judged, it is achieved that the tightening process is not limited by fields and environments, no torque sensor is used in the monitoring process, torque information and the number of torsion turns can be monitored, and the abnormal condition is fed back to an operator immediately.

Description

technical field [0001] The invention relates to a sample collection system, a deep learning network and a monitoring system in an assembly tightening process, and belongs to the field of assembly monitoring. Background technique [0002] Mechanical assembly is an important part of the machinery manufacturing industry. It is the process of realizing the combination of mechanical parts and components and completing the machine assembly according to the technical requirements. During the assembly process, if the bolts and other connecting parts are not tightened, the connection relationship between the two parts will be affected, and the assembly quality and assembly efficiency will be affected. [0003] Monitoring the assembly and tightening process can detect problems in manual assembly in time by monitoring the tightening force, torque and other factors on the connecting parts in real time, thereby improving the quality of product installation. Especially in customized prod...

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): G06K9/00G06N3/08
CPCG06N3/084G06V20/41G06F2218/12
Inventor 黄凯陈成军李东年洪军
Owner QINGDAO TECHNOLOGICAL UNIVERSITY
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