基于改进ViT的熔池识别与焊接偏差在线检测方法*

doi:10.16731/j.cnki.1671-3133.2024.10.017

现代制造工程 ›› 2024, Vol. 529 ›› Issue (10): 130-137.doi: 10.16731/j.cnki.1671-3133.2024.10.017

• 仪器仪表/检测/监控 • 上一篇下一篇

基于改进ViT的熔池识别与焊接偏差在线检测方法^*

蒋宇轩¹, 林凯¹, 王瑶祺¹, 张岳¹, 洪宇翔^1,2

1 中国计量大学机电工程学院,杭州 310018;
2 中国计量大学浙江省智能制造质量大数据溯源与应用重点实验室,杭州 310018

收稿日期:2024-02-02 发布日期:2024-10-29
通讯作者: 洪宇翔,博士,副教授,硕士研究生导师,主要研究方向为智能化焊接与增材制造。E-mail:hongyuxiang@cjlu.edu.cn
作者简介:蒋宇轩,硕士研究生,主要研究方向为机器人智能化焊接。
基金资助:
^*国家自然科学基金项目(51605251);浙江省自然科学基金项目(LY22E050009);浙江省教育厅科研资助项目(Y202249427);浙江省教育厅科研资助项目(Y202147838);浙江省属高校基本科研业务费专项资金资助项目(2023YW41)

Improved ViT-based method for molten pool recognition and online detection of welding deviation

JIANG Yuxuan¹, LIN Kai¹, WANG Yaoqi¹, ZHANG Yue¹, HONG Yuxiang^1,2

1 College of Mechanical and Electrical Engineering,China Jiliang University,Hangzhou 310018,China;
2 Key Laboratory of Intelligent Manufacturing Quality Big Data Tracing and Analysis of Zhejiang Province, China Jiliang University,Hangzhou 310018,China

Received:2024-02-02 Published:2024-10-29

摘要/Abstract

摘要： 焊接偏差的精确检测是实现焊接机器人焊缝轨迹自动跟踪及智能化焊接的前提。提出了一种基于改进视觉转换器(Vision Transformer,ViT)的熔池识别与焊接偏差在线检测方法。首先,采用轻量级ViT模型Segformer作为基线模型,在其掩码分割前嵌入置换注意力(Shuffle Attention,SA)机制,以更好地捕获特征信息在空间和通道这2个维度中的依赖关系,从而提高模型的分割精度;其次,在多层感知机(Multilayer Perceptron,MLP)中加入上下文广播(Context Broadcasting,CB)模块,在保证模型低参数量的前提下提高泛化能力;最后,基于模型分割结果,提出一种焊接偏差计算方法来定量描述偏差检测精度。实验结果表明,相较于基线模型,所提出模型的平均交并比和平均像素准确率分别提高了2.67 %和2.12 %,且对于不同预设焊枪偏移情况均具有良好的泛化性,焊接偏差精度控制在±0.021 mm之内,为实现精密焊接焊缝跟踪提供基础。

关键词: 焊接偏差, 焊缝跟踪, 熔池识别, 视觉转换器, 注意力机制

Abstract: Accurate detection of welding deviations is a prerequisite for automatic seam tracking and intelligent welding by welding robots.An improved ViT-based method for molten pool recognition and online detection of welding deviation was proposed. Firstly,the lightweight ViT model Segformer was used as the baseline model. The Shuffle Attention (SA) was embedded before mask segmentation to better capture the dependencies of feature information in both spatial and channel dimensions. Thus,the model's segmentation accuracy was enhanced.Secondly,a Context Broadcasting (CB) module was added to the Multilayer Perceptron (MLP) to improve the generalization capability while ensuring low parameters of model. Finally,based on the model segmentation results,a welding deviation calculation method was proposed to quantitatively describe the deviation detection accuracy. The experimental results show that,compared with the baseline model,the mean intersection over union and mean pixel accuracy of proposed model were increased by 2.67 % and 2.12 %,respectively,and it has good generalization for different preset torch offsets. The welding deviation accuracy was controlled between ±0.021 mm,which provided a basis for seam tracking in precision welding.

Key words: welding deviation, seam tracking, molten pool recognition, Vision Transformer (ViT), attention mechanism

中图分类号:

TG409

蒋宇轩, 林凯, 王瑶祺, 张岳, 洪宇翔. 基于改进ViT的熔池识别与焊接偏差在线检测方法^*[J]. 现代制造工程, 2024, 529(10): 130-137.

JIANG Yuxuan, LIN Kai, WANG Yaoqi, ZHANG Yue, HONG Yuxiang. Improved ViT-based method for molten pool recognition and online detection of welding deviation[J]. Modern Manufacturing Engineering, 2024, 529(10): 130-137.

参考文献

[1] ZHANG Y,WANG Q.Adaptive Intelligent Welding Manufacturing[J].Welding Journal,2021,100(1):63-83.
[2] LEE S K.A Study on Automatic Seam Tracking in Pulsed Laser Edge Welding by Using a Vision Sensor Without an Auxiliary Light Source[J].Journal of Manufacturing Systems,2002,21(4):302-315.
[3] 郭岩宝,王斌,王德国,等.焊接机器人的研究进展与发展趋势[J].现代制造工程,2021(5):53-63.
[4] HE Y,MA G,CHEN S,et al.Autonomous Decision-Making of Welding Position During Multipass GMAW With T-Joints:A Bayesian Network Approach[J].IEEE Transactions on Industrial Electronics,2022,69(4):3909-3917.
[5] HONG Y,CHANG B,PENG G,et al.In-Process Monitoring of Lack of Fusion in Ultra-Thin Sheets Edge Welding Using Machine Vision[J].Sensors,2018,18(8):2411.
[6] GUO B,SHI Y,YU G,et al.Weld deviation detection based on wide dynamic range vision sensor in MAG welding process[J].The International Journal of Advanced Manufacturing Technology,2016,87(9/10/11/12):3397-3410.
[7] ZHU J,WANG J,SU N,et al.An infrared visual sensing detection approach for swing arc narrow gap weld deviation[J].Journal of Materials Processing Technology,2017,243:258-268.
[8] BAE K Y,LEE T H,AHN K C,et al.An optical sensing system for seam tracking and weld pool control in gas metal arc welding of steel pipe[J].Journal of Materials Processing Technology,2002,120(1/2/3):458-465.
[9] XU Y,LÜ N,FANG G,et al.Welding seam tracking in robotic gas metal arc welding[J].Journal of Materials Processing Technology,2017,248:18-30.
[10] ZHANG B,SHI Y,GU S,et al.Narrow-seam identification and deviation detection in keyhole deep-penetration TIG welding[J].The International Journal of Advanced Manufacturing Technology,2019,101(5/6/7/8):2051-2064.
[11] CHEN Y,SHI Y,CUI Y,et al.Narrow gap deviation detection in Keyhole TIG welding using image processing method based on Mask-RCNN model[J].The International Journal of Advanced Manufacturing Technology,2021,112(7):2015-2025.
[12] LIN Z,SHI Y,WANG Z,et al.Intelligent Seam Tracking of an Ultranarrow Gap During K-TIG Welding:A Hybrid CNN and Adaptive ROI Operation Algorithm[J].IEEE Transactions on Instrumentation and Measurement,2023,72:1-14.
[13] XIE E,WANG W,YU Z,et al.SegFormer:Simple and efficient design for semantic segmentation with transformers[J].Advances in Neural Information Processing Systems,2021,34.
[14] ZHANG Q L,YANG Y B.SA-Net:Shuffle Attention for Deep Convolutional Neural Networks[C]//Internation Conference an Aeoustics, Speech and Siqnal Processing(ICASSP).[S.l.]:IEEE,2021.
[15] HYEON-WOO N,YU-JI K,HEO B,et al.Scratching Visual Transformer’s Back with Uniform Attention[J].ArXiv,2022,10(16).
[16] RONNEBERGER O,FISCHER P,BROX T.U-Net:Convolutional Networks for Biomedical Image Segmentation[C]//Navab N,Hornegger J,Wells W M,et al.Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015.Cham:Springer International Publishing,2015:234-241.
[17] CHEN L C,ZHU Y,PAPANDREOU G,et al.Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation[C]//FERRARI V,HEBERT M,SMINCHISESCU C,et al.Computer Vision-ECCV 2018.Cham:Springer International Publishing,2018:833-851.
[18] ZHAO H,SHI J,QI X,et al.Pyramid Scene Parsing Network[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Honolulu,HI:IEEE,2017:6230-6239.

基于改进ViT的熔池识别与焊接偏差在线检测方法^*

Improved ViT-based method for molten pool recognition and online detection of welding deviation

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 11

编辑推荐

Metrics

[1]	孙铁强, 秦愿伟, 宋超, 肖鹏程. 基于改进YOLOv7的钢卷端面缺陷检测^*[J]. 现代制造工程, 2024, 526(7): 117-125.
[2]	马涛, 李敬兆. 基于MFA-UNet的铜制螺纹零件外表面缺陷检测^*[J]. 现代制造工程, 2024, 524(5): 113-120.
[3]	莫坚;张泽. 轴承剩余使用寿命的注意力多尺度卷积神经网络预测[J]. 现代制造工程, 2023, 515(8): 148-154.
[4]	王瑞，李敬兆，石晴，许志，王龙，赵雪琦，陈强. 基于融合多维非对称策略的螺纹缺陷检测方法[J]. 现代制造工程, 2023, 513(6): 107-116.
[5]	阎馨，杨月川，屠乃威. 基于改进SSD的钢材表面缺陷检测[J]. 现代制造工程, 2023, 512(5): 112-120.
[6]	王凯，吉卫喜，卢璟钰，苏璇. 基于改进GADF-SeResNet滚动轴承故障诊断方法[J]. 现代制造工程, 2023, 512(5): 135-142.
[7]	薛阳;丁凯;李清;杨江天;李金星. 基于改进YOLOv5s的活塞杆表面缺陷检测[J]. 现代制造工程, 2023, 518(11): 104-112.
[8]	仇娇慧，贝绍轶，尹明锋，卿宏军. 基于改进YOLOv5s的齿轮表面缺陷检测[J]. 现代制造工程, 2022, 498(3): 104-113.
[9]	黄昌顺，张金萍. 基于CBAM-CNN的滚动轴承故障诊断方法[J]. 现代制造工程, 2022, 506(11): 137-143.
[10]	党顺峰, 熊锐, 李继辉, 陈灿奇, 陈振威, 吴鑫. 基于改进Mask-RCNN算法的车位检测研究[J]. 现代制造工程, 2021, 484(1): 91-97.
[11]	王凯，孙磊，何宏. 一种基于双相机焊缝跟踪的球形储罐爬壁机器人定位方法[J]. 现代制造工程, 2020, 476(5): 15-24.