Welding trajectory real-time detecting method in welding process of welding robot

A welding robot and welding process technology, applied in welding equipment, welding equipment, auxiliary welding equipment, etc., can solve the problems of complex on-site installation and high cost, and achieve the effect of simple and fast implementation, convenient maintenance, and simple and fast installation.

Active Publication Date: 2019-09-17
SHANGHAI ZHANWAN INFORMATION SCI & TECHCO LTD
8 Cites 6 Cited by

AI-Extracted Technical Summary

Problems solved by technology

The cost of this solution is relatively high, and the instal...
View more

Abstract

A welding trajectory real-time detecting method in the welding process of a welding robot includes the steps that welding trajectory data in the welding process of a normal welding robot are acquired; dimensionality reduction analysis is performed on the acquired welding trajectory data, and the cluster number of distribution dense clusters of the data is determined; according to the cluster data, all data points are accurately classified, and center point coordinates of each category are obtained; most TCP data point coordinate values of each category are substituted into a BP neutral network algorithm for network training; remaining TCP coordinate values are substituted into a neutral network algorithm for network authentication success; welding trajectory data of a certain point are acquired in real time; TCP-X and TCP-Y coordinate values in actual TCP coordinate values are substituted into the neutral network algorithm so that a predicted TCP-Z coordinate value can be obtained; and a difference between the predicted TCP-Z coordinate value and an actual TCP-Z coordinate value are is calculated, the point is a normal point if the difference falls into a preset confidence interval, and the point is an abnormal point if the difference does not fall into the preset confidence interval, and predicting alarming is performed.

Application Domain

Programme-controlled manipulatorWelding/cutting auxillary devices +3

Technology Topic

Real time acquisitionData point +10

Image

  • Welding trajectory real-time detecting method in welding process of welding robot

Examples

  • Experimental program(1)

Example Embodiment

[0028] In order to make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be described clearly and completely in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of the embodiments of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present invention.
[0029] Such as figure 1 As shown, this embodiment provides a method for real-time detection of welding trajectories during welding by a welding robot, which includes the following steps:
[0030] Step 101: Collect welding trajectory data in a three-dimensional space during a normal welding robot welding process. The welding trajectory data includes weld number and TCP coordinate values, and TCP coordinate values ​​include TCP-X, TCP-Y and TCP-Z coordinate values.
[0031] Step 102: Use a principal component analysis (PCA) algorithm to perform dimensionality reduction analysis on the collected three-dimensional welding trajectory data. On the basis of almost no loss of data distribution characteristic information, the data is projected into a low-dimensional space (for example, 2D or 1D) through appropriate spatial rotation for visual analysis. Roughly determine the number of data distribution categories, that is, the number of clusters (k) of densely distributed clusters.
[0032] Step 103: According to the cluster number k in the PCA process, use the Kmeans clustering algorithm to accurately classify all data points, and obtain the center point coordinates of each class.
[0033] Step 104: Substitute most of the TCP data point coordinate values ​​of each category into the BP neural network algorithm for network training, and alternately:
[0034] Output (TCP-Z)
[0035] Output (TCP-X)
[0036] Output (TCP-Y)
[0037] As the input and output of the BP neural network algorithm, iteratively trains multiple times to find the best matching method and the model formula obtained.
[0038] Step 105: Substitute the remaining TCP coordinate values ​​into the BP neural network algorithm for network verification, and the verification is successful.
[0039] Step 106: Collect the welding track data of a certain point in the welding process of a welding robot in real time.
[0040] Step 107: Substitute the TCP-X and TCP-Y coordinate values ​​of the actually collected TCP coordinate values ​​into the BP neural network algorithm to obtain the predicted TCP-Z coordinate values.
[0041] Step 108: Calculate the gap between the predicted TCP-Z coordinate value and the actual TCP-Z coordinate value. If the gap falls within the preset confidence interval, the point is a normal point, and the gap does not fall within the preset confidence interval , Then the point is an abnormal point and a predictive warning is performed.
[0042] Alternatively, in step 107, the TCP-X and TCP-Z coordinate values ​​in the actually collected TCP coordinate values ​​are substituted into the neural network algorithm to obtain the predicted TCP-Y coordinate values.
[0043] Step 108: Calculate the gap between the predicted TCP-Y coordinate value and the actual TCP-Y coordinate value. If the gap falls within the preset confidence interval, the point is a normal point, and the gap does not fall within the preset confidence interval , Then the point is an abnormal point and a predictive warning is performed.
[0044] Alternatively, in step 107, the TCP-Y and TCP-Z coordinate values ​​in the actually collected TCP coordinate values ​​are substituted into the neural network algorithm to obtain the predicted TCP-X coordinate values.
[0045] Step 108: Calculate the gap between the predicted TCP-X coordinate value and the actual TCP-X coordinate value. If the gap falls within the preset confidence interval, the point is a normal point, and the gap does not fall within the preset confidence interval , Then the point is an abnormal point and a predictive warning is performed.
[0046] Because the trend of the data becomes irregular, it is decided to adopt a neural network-based non-parametric deep learning fitting method and training data, and then based on the deviation vector obeys the normal distribution law combined with the quantile distance IQR abnormal point detection method to establish upper and lower confidence Interval.
[0047] And calculate the gap between the actual value and the predicted value. If the gap falls within the confidence interval, the point is regarded as a normal point. If it falls outside the confidence interval, it is an abnormal point and an alarm is generated and sent to the platform.
[0048] The invention combines the welding trajectory data collected in real time, trains and learns the welding trajectory of the robot through kmeans clustering, neural network algorithm, etc. to obtain the distribution law of the welding trajectory; real-time detection of welding deviation defects: theoretical welding based on big data analysis Seam model, real-time detection of current welding whether there is a welding deviation defect.
[0049] Although the specific embodiments of the present invention are described above, those skilled in the art should understand that these are only examples, and the protection scope of the present invention is defined by the appended claims. Those skilled in the art can make various changes or modifications to these implementations without departing from the principle and essence of the present invention, but these changes and modifications all fall within the protection scope of the present invention.

PUM

no PUM

Description & Claims & Application Information

We can also present the details of the Description, Claims and Application information to help users get a comprehensive understanding of the technical details of the patent, such as background art, summary of invention, brief description of drawings, description of embodiments, and other original content. On the other hand, users can also determine the specific scope of protection of the technology through the list of claims; as well as understand the changes in the life cycle of the technology with the presentation of the patent timeline. Login to view more.

Similar technology patents

Experimental device for self-adaption structure teaching

ActiveCN103456207AStimulate interest in learningEasy and quick installation
Owner:NANJING UNIV OF TECH

Convenient single-sided three-dimensional adjusting door window hinge

InactiveCN110952858ANovel and perfect structureEasy and quick installation
Owner:JIANGXI FENFA TECH +1

Steel strand fixing device

InactiveCN104652709ASimple structureEasy and quick installation
Owner:天津市力胜通预应力工程有限公司

Environment-friendly efficient purification faucet

InactiveCN108488475AEasy and quick installationavoid accidental damage
Owner:傅华中

Illuminable ultraviolet disinfection wall hanging lamp

InactiveCN108478820AEasy and quick installation
Owner:SUZHOU CHENGTENG ELECTRONICS TECH

Classification and recommendation of technical efficacy words

  • Easy and quick installation
  • Simple and quick to implement

Method for mounting large-area refractory fiber module

ActiveCN102425949AEasy and quick installationgood appearance quality
Owner:CHINA FIRST METALLURGICAL GROUP

Flexible coil current transformer and production method thereof

InactiveCN102751081AEasy and quick installation
Owner:SHENZHEN CHUANGYIN TECH

Outdoor self-locking LED display screen

ActiveCN106409166AEasy and quick installationeasy maintenance
Owner:浙江德广信电子科技股份有限公司

Gas turbine generator

InactiveCN1877969AEasy and quick installationadvanced technology
Owner:SICHUAN DONGFENG ELECTRIC MACHINARY WORKS CO LTD

Emergency heating method of emergency heating device for portable SF6 tank type circuit breaker

InactiveCN107749362AEasy and quick installationSimple structure
Owner:国网黑龙江省电力有限公司超高压公司 +1

Coke oven checker brick mounting device and construction method therefor

ActiveCN110144231ASimple and quick to implementImprove work efficiency
Owner:CHINA FIRST METALLURGICAL GROUP

Soft tent

PendingCN108839545ASimple and quick to implementeasy to use
Owner:DONGGUAN LONMAN AUTO & CAMPING ACCESSORIES MFG
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