Infrared Visual Sensing Detection Approach of Swing Arc Narrow Gap Weld Deviation Based on Outlier Data Filtering

Su, Na; Wang, Jie; Xu, Guoxiang; Zhu, Jie; Wang, Jiayou

doi:10.1109/jsen.2022.3189681

Cited by 5 publications

(2 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, since these studies were conducted based on a specific form, their application is limited in unstructured conditions where predefined forms are not set. An arc-sensing study has also been conducted for welding in a narrow gap, and shares similarities with this study in that it sets arc-sensing conditions for workpieces under challenging conditions [31]. However, this study is limited to very specific conditions, making it difficult to apply arc sensing to workpieces with gaps between them or workpieces with curvature, which are common in shipyards.…”

Section: Related Workmentioning

confidence: 97%

Data Clustering Utilization Technologies Using Medians of Current Values for Improving Arc Sensing in Unstructured Environments

Kim,

Heo

et al. 2024

Sensors

View full text Add to dashboard Cite

In the shipbuilding industry, welding automation using welding robots often relies on arc-sensing techniques due to spatial limitations. However, the reliability of the feedback current value, core sensing data, is reduced when welding target workpieces have significant curvature or gaps between curved workpieces due to the control of short-circuit transition, leading to seam tracking failure and subsequent damage to the workpieces. To address these problems, this study proposes a new algorithm, MBSC (median-based spatial clustering), based on the DBSCAN (density-based spatial clustering of applications with noise) clustering algorithm. By performing clustering based on the median value of data in each weaving area and considering the characteristics of the feedback current data, the proposed technique utilizes detected outliers to enhance seam tracking accuracy and responsiveness in unstructured and challenging welding environments. The effectiveness of the proposed technique was verified through actual welding experiments in a yard environment.

show abstract

Section: Related Workmentioning

confidence: 97%

Data Clustering Utilization Technologies Using Medians of Current Values for Improving Arc Sensing in Unstructured Environments

Kim,

Heo

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…The use of robot welding than manual welding accuracy stability is higher. Literature [14] proposes the outlier data filtering (ODF) method to detect the deviation of the weld seam and weld pitch center. The method creates a notch center detection algorithm consisting of global pattern recognition as the underlying framework and a position consistency filtering (PCF) algorithm based on optimal estimation to do the localization.…”

Section: Introductionmentioning

confidence: 99%

Research Progress of Weld Tracking Image Processing Technology Based on Deep Learning Theory

Shen,

2023

Applied Mathematics and Nonlinear Sciences

View full text Add to dashboard Cite

In this paper, a convolutional neural network is used to localize the weld seam feature points with noise interference in complex welding environments. A priori frames are introduced into the feature point extraction network, combined with position prediction and confidence prediction, to improve the accuracy and anti-interference ability of the weld tracking system. To improve welding efficiency by utilizing the continuity of weld tracking, the weld tracking network is designed based on the twin structure. The weld detection network designs the first frame to locate the key position of the bevel and inputs into the weld tracking network as a template, and the weld tracking network completes the automatic tracking of the subsequent welds. At the same time, the network introduces a hybrid domain attention mechanism, which makes full use of the weld feature channel dependence and spatial location relationship and puts more attention near the inflection point of the weld laser line to ensure the accuracy of weld tracking. The research results show that the extraction error of weld seam feature points based on the convolutional neural network is within 17, which is much lower than that of the grayscale center of gravity method and Steger's algorithm. In the weld tracking experiments under the workpiece tilting state, the average value of the absolute error of the tracking trajectory in the X-axis direction is not more than 0.7 mm, and the maximum value is less than 1.15 mm. The absolute tracking error in the Z-axis direction is relatively low, with an average of 0.638 mm and a maximum of 1.573 mm. Therefore, the weld-tracking image processing technique proposed in this paper has strong anti-noise interference capabilities and high localization accuracy. And high accuracy in localization.

show abstract

Thick-wire swing arc narrow gap GMA welding assisted by pre-embedding cold wires

Zhu,

Wang,

Liu

et al. 2024

Int J Adv Manuf Technol

Self Cite

View full text Add to dashboard Cite

To raise narrow gap GMA welding efficiency while ensuring weld formation, the present work presents a thick wire swing arc narrow gap GMA welding process assisted by cold wires. This process employs the thick wire and the pre-embedded cold wires to jointly increase the welding efficiency, while improving the groove sidewall penetration by the swing arc. Two types of thick wires, being conventional single-strand welding wire (SWW) and cable-type welding wire (CWW), are respectively used as main wire. After solving the series arc problem during CWW arc-igniting, the effects of the pre-embedded wire layer number on narrow gap weld formation and welding efficiency are investigated for the thick wire welding. Experimental results show that the weld thickness and the sidewall penetration do not obviously change with increasing the number of the cold wires, while the weld layer height significantly grows. Additionally, the SWW weld formation looks more sound. Compared to the thin-wire welding, the thick wire welding without the cold wire raises the deposition rate by about 38% and 58% as respectively using SWW and CWW. When pre-embedding two layers of cold wires, the increased percentage of the deposition rate further rises respectively to about 81% and 112%. These thus demonstrate the effectiveness of the developed process in addition to the higher efficiency of the CWW welding.

show abstract

Infrared Visual Sensing Detection Approach of Swing Arc Narrow Gap Weld Deviation Based on Outlier Data Filtering

Cited by 5 publications

References 27 publications

Data Clustering Utilization Technologies Using Medians of Current Values for Improving Arc Sensing in Unstructured Environments

Data Clustering Utilization Technologies Using Medians of Current Values for Improving Arc Sensing in Unstructured Environments

Research Progress of Weld Tracking Image Processing Technology Based on Deep Learning Theory

Thick-wire swing arc narrow gap GMA welding assisted by pre-embedding cold wires

Contact Info

Product

Resources

About