A Deep Learning Fusion Model of Wave and Image Data for Weld Defect Recognition

doi:10.7652/xjtuxb202105009

您当前的位置：

首页 >

文章列表页 >

A Deep Learning Fusion Model of Wave and Image Data for Weld Defect Recognition

更新时间：2025-07-09

- A Deep Learning Fusion Model of Wave and Image Data for Weld Defect Recognition
- Vol. 55, Issue 5, Pages: 73-82(2021)
- 作者机构：
  
  1. 西安交通大学机械制造系统工程国家重点实验室,西安,710049
  2. 陕西省特种设备检验检测研究院,西安,710048
- 作者简介：
- 基金信息：
- DOI：10.7652/xjtuxb202105009
  CLC： TP391
- Online First：10 May 2021，
  
  Published：2021
- 稿件说明：
移动端阅览
A Deep Learning Fusion Model of Wave and Image Data for Weld Defect Recognition[J]. 2021, 55(5): 73-82.
DOI：

A Deep Learning Fusion Model of Wave and Image Data for Weld Defect Recognition[J]. 2021, 55(5): 73-82. DOI： 10.7652/xjtuxb202105009.

摘要

针对当前利用超声衍射时差法(TOFD)图谱数据进行缺陷类型判定时以人工判别为主

主观性大、效率低以及缺乏从波形特征及图像特征进行集成分析的问题

提出了一种深度学习融合模型及焊缝缺陷识别方法。通过对TOFD检测原理及缺陷检测图谱数据特点进行分析

建立了综合考虑波形数据和图像数据的缺陷特征表征方法

实现了缺陷标准数据集构建; 通过构建基于时间卷积网络(TCN)的波形序列数据分析模块、基于卷积神经网络(CNN)的图像数据分析模块以及特征自适应融合分类模块

建立了一种可以实现波形序列特征与图像特征综合分析的深度学习融合网络模型(DLFM)及模式分类方法。以企业实际TOFD检测焊缝缺陷数据对所提方法进行了验证

结果表明所提DLFM方法对缺陷类型的识别率明显高于单独使用基于TCN、CNN以及CNN-TCN的方法; 所提方法拓展了现有深度学习模型的构建方法

并可以推广应用到其他模式识别领域

具有较强的通用性。

Abstract

Aiming at the problems of manual identification

subjectivity

low efficiency

and lack of integrated analysis from waveform features and image features in using the time-of-flight

diffraction(TOFD)data for weld defect recognition

a deep learning fusion model(DLFM)and a weld defect recognition method are proposed. Based on the analyses of TOFD detection principle and weld defect detection data characteristics

a defect feature representation method considering waveform data and image data is set up

and the defect standard data set is established. Combining waveform sequence data analysis module based on time convolution network(TCN)

image data analysis module based on convolutional neural network(CNN)and feature adaptive fusion classification module

the DLFM with pattern classification is constructed. A case study of TOFD weld defect recognition is conducted to illustrate the work. The results show that the proposed DLFM has higher defect recognition rate than the method based on CNN

TCN or CNN-TCN. The proposed method improves the traditional deep learning models

and can be applied to the other pattern recognition fields with stronger universality.

关键词

Keywords

references

BABY S, BALASUBRAMANIAM T, PARDIKAR R J, et al. Time-of-flight diffraction technique for accurate sizing of surface-breaking cracks [J]. Insight-Non-Destructive Testing and Condition Monitoring, 2003, 45(6): 426-430.

BASKARAN G, BALASUBRAMANIAM K, LAKSHMANA RAO C. Shear-wave time of flight diffraction(S-TOFD)technique [J]. NDT and E International, 2006, 39(6): 458-467.

PRAVEEN A, VIJAYAREKHA K, ABRAHAM S T, et al. Fourier analysis of ultrasonic TOFD signals for defect detection in austenitic stainless steel welds [J]. International Journal of Computer Applications, 2013, 71(9): 14-17.

陈尧, 冒秋琴, 石文泽, 等. 基于相位相干性的厚壁焊缝TOFD成像检测研究 [J]. 机械工程学报, 2019, 55(4): 25-32.

CHEN Yao, MAO Qiuqin, SHI Wenze, et al. Research on ultrasonic TOFD imaging inspection for heavy-walled weld based on phase coherence characteristics [J]. Journal of Mechanical Engineering, 2019, 55(4): 25-32.

SILVA L C, FILHO E F S, ALBUQUERQUE M C S, et al. Segmented analysis of time-of-flight diffraction ultrasound for flaw detection in welded steel plates using extreme learning machines [J]. Ultrasonics, 2019, 102: 106057.

盛朝阳, 刚铁, 迟大钊. 基于分水岭方法的超声TOFD检测图像分割 [J]. 机械工程学报, 2011, 47(8): 35-40.

SHENG Zhaoyang, GANG Tie, CHI Dazhao. Segmentation of ultrasonic TOFD testing image based on watershed method [J]. Journal of Mechanical Engineering, 2011, 47(8): 35-40.

陈振华, 胡怀辉, 卢超. 基于超声TOFD直通波及神经网络的近表面缺陷自动识别 [J]. 无损检测, 2014, 36(3): 14-17.

CHEN Zhenhua, HU Huaihui, LU Chao. Automatic identification technology of near surface defects based on neural network and through wave of ultrasonic TOFD [J]. Nondestructive Testing, 2014, 36(3): 14-17.

JIN S J, SUN X, MA T T, et al. Quantitative detection of shallow subsurface defects by using mode-converted waves in time-of-flight diffraction technique [J]. Journal of Nondestructive Evaluation, 2020, 39(2): 1-8.

姜洪权, 贺帅, 高建民, 等. 一种改进卷积神经网络模型的焊缝缺陷识别方法 [J]. 机械工程学报, 2020, 56(8): 235-242.

JIANG Hongquan, HE Shuai, GAO Jianmin, et al. An improved convolutional neural network for weld defect recognition [J]. Journal of Mechanical Engineering, 2020, 56(8): 235-242.

黄焕东, 胡利晨, 李斌彬, 等. 基于区域的快速卷积神经网络的焊缝TOFD检测缺陷识别 [J]. 无损检测, 2019, 41(7): 12-18.

HUANG Huandong, HU Lichen, LI Binbin, et al. Recognition of defect in TOFD image based on faster region convolutional neural networks [J]. Nondestructive Testing, 2019, 41(7): 12-18.

HAMMAD M, LIU Yashu WANG Kuanquan. Multimodal biometric authentication systems using convolution neural network based on different level fusion of ECG and fingerprint [J]. IEEE Access, 2019, 7: 25527-25542.

MARKUS K, FRIEDHELM S. Cascaded fusion of dynamic, spatial, and textural feature sets for person-independent facial emotion recognition [C]∥Proceedings of the 22nd International Conference on Pattern Recognition. Piscataway, NJ, USA: IEEE Computer Society, 2014: 4660-4665.

吕磊, 张艳飞, 谢利明, 等. TOFD检测技术应用和发展 [J]. 内燃机与配件, 2020(7): 246-248.

LÜ Lei, ZHANG Yanfei, XIE Liming, et al. Application and development of TOFD detection technology [J]. Internal Combustion Engine and Parts, 2020(7): 246-248.

LÉCUN Y, BOTTOU L, BENGIO Y, et al. Gradient-based learning applied to document recognition [J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.

BAI Shaojie, ZICO KOLTER J, KOLTUN V. An empirical evaluation of generic convolutional and recurrent networks for sequence modeling [EB/OL].(2018-03-04)[2020-08-06]. https:∥arxiv.org/abs/1803. 01271.

GREFF K, SRIVASTAVA R K, KOUTNÍK J, et al. LSTM: a search space odyssey [J]. IEEE Transactions on Neural Networks and Learning Systems, 2017, 28(10): 2222-2232.

AL-JANABI S, MOHAMMAD M, AL-SULTAN A. A new method for prediction of air pollution based on intelligent computation [J]. Soft Computing, 2020, 24(1): 661-680.

HANG Renlong, LIU Qingshan, HONG Danfeng. Cascaded recurrent neural networks for hyperspectral image classification [J]. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57(8): 5384-5394.

WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module [C]∥Proceedings of the 15th European Conference on Computer Vision. Berlin, Germany: Springer, 2018: 3-19.

李伟, 杨向东, 陈恳. 基于CNN和RNN联合网络的心音自动分类 [J]. 计算机工程与设计, 2020 41(1): 46-51.

LI Wei, YANG Xiangdong, CHEN Ken. Heart sound classification based on CNN and RNN [J]. Computer Engineering and Design, 2020 41(1): 46-51.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Intelligent Quantitative Non-Destructive Evaluation Method for Delamination of Carbon Fiber Composite Tube with Mirror-Reflection Infrared Thermography

Model Predictive Control-Oriented Deep Koopman Global Linear Modeling Method for Wind Turbines

Dynamic Liquid Level Prediction of Steam Generator under Main Steam Pipe Rupture Accidents

A Pure Convolutional Model Fused with U-net Network for Video Prediction

Application of Interpretable Convolutional Neural Network Enabled by Bi-Damped Wavelets in Bearing Fault Diagnosis

Related Author

YANG Yang

CHEN Hongen

LIU Bofan

TONG Zongfei

PEI Cuixiang

CHEN Zhenmao

TIAN Runze

KOU Peng

Related Institution

School of Aerospace Engineering，Xi’an Jiaotong University

State Key Laboratory for Strength and Vibration of Mechanical Structures，Xi’an Jiaotong University

Shaanxi Engineering Research Center of Nondestructive Testing and Structural Integrity Evaluation，Xi’an Jiaotong University

Northern Branch, Beijing Jingneng International Holding Co., Ltd.

State Key Laboratory of Electrical Insulation and Power Equipment, Xi'an Jiaotong University

⁰