Weld defect detection based on adaptive fusion of multi-domain and multi-scale deep features

被引：0

作者：

Zhang R. ^{[1
,2
]}

Gao M. ^{[1
,2
]}

Fu L. ^{[1
,2
]}

Zhang P. ^{[1
]}

Bai X. ^{[1
]}

Zhao N. ^{[1
,2
]}

机构：

[1] College of Computer Science and Technology, Taiyuan University of Science and Technology, Taiyuan

[2] Shanxi Electromechanical Design and Research Institute Co., Ltd., Taiyuan

来源：

Zhendong yu Chongji/Journal of Vibration and Shock | 2023年 / 42卷 / 17期

关键词：

convolutional neural network (CNN) model optimization strategy; model self-optimization; multi-domain and multi-scale feature fusion; ultrasonic detection; weld defect;

D O I：

10.13465/j.cnki.jvs.2023.17.036

中图分类号：

学科分类号：

摘要：

Here, aiming at problems of low information richness of weld defect detection signals and strong manual dependence on deep network architecture, weld defect detection studies were performed based on adaptive fusion of multi-domain and multi-scale deep features. Firstly, a time-domain dataset was constructed and derived into real and complex domains to enrich feature expression of detection signals. Secondly, a multi-domain information fusion model was designed to fully integrate feature domain information. Finally, a model optimization strategy oriented to multi-dimensional hyperparametric self-optimization of convolutional neural network (CNN) was proposed to improve the model' s efficiency and performance. Tests showed that the proposed method has an accuracy of 96. 54% for identifying 5 types of weld defects; it can improve the recognition accuracy while maintaining a smaller number of parameters and computational consumption; it has stronger practicality and generalization. © 2023 Chinese Vibration Engineering Society. All rights reserved.

引用

下载

页码：294 / 305+313

共 24 条

[1] WANG Rui, HU Yunlei, LIU Weipeng, Et al., Weld defect detection based on edge AI in X-ray image [J], Transactions of the China Welding Institution, 43, 1, pp. 79-84, (2022)
[2] AJMI C, ZAPATA J, MARTINEZ-ALVAREZ J J, Et al., Using deep learning for defect classification on a small weld X-ray image dataset [J], Journal of Nondestructive Evaluation, 39, 3, pp. 1-13, (2020)
[3] HUANG Huandong, HU Lichen, LI Binbin, Et al., Recognition of defect in TOFD image based on faster region convolutional neural networks [J], Nondestructive Testing, 41, 7, pp. 12-18, (2019)
[4] HU Hongwei, ZHANG Jie, PENG Gang, Et al., Defect classification method for ultrasonic inspection of welding seams based on LBP-KPCA feature extraction [J], Transactions of the China Welding Institution, 40, 6, pp. 34-39, (2019)
[5] WANG X, GUAN S, HUA L, Et al., Classification of spot-welded joint strength using ultrasonic signal time-frequency features and PSO-SVM method, Ultrasonics, 91, pp. 161-169, (2019)
[6] MURTA R, VIEIRA F, SANTOS V, Et al., Welding defect classification from simulated ultrasonic signals [J], Journal of Nondestructive Evaluation, 37, 3, (2018)
[7] SILVA L C, FILHO E, ALBUQUERQUE M, Et al., Segmented analysis of time-of-flight diffraction ultrasound for flaw detection in welded steel plates using extreme learning machines, Ultrasonics, 102, 11, (2019)
[8] ZHANG Xunjie, ZHANG Min, LI Xianjun, Rolling bearing fault mode recognition based on 2D image and CNN-BiGRU [J], Journal of Vibration and Shock, 40, 23, pp. 194-201, (2021)
[9] LU L, WANG Z., Encoding temporal Markov dynamics in graph for time series visualization, (2016)
[10] MENG X J, QIU S, WAN S, Et al., A motor imagery EEG signal classification algorithm based on recurrence plot convolution neural network [J], Pattern Recognition Letters, 146, 4, pp. 134-141, (2021)

← 1 2 3 →