基于卷积神经网络的涡流热成像检测技术

doi:10.3969/j.issn.1000-1565.2025.06.011

河北大学学报(自然科学版) ›› 2025, Vol. 45 ›› Issue (6): 653-660.DOI: 10.3969/j.issn.1000-1565.2025.06.011

• • 上一篇

基于卷积神经网络的涡流热成像检测技术

李建宇^1,2,3,程姣姣^1,2,朱和明^1,2,郑文培³

收稿日期:2023-11-01 发布日期:2025-11-19
作者简介:李建宇(1988—),男,国家页岩油气富集机理与有效开发重点实验室助理研究员,博士,主要从事无损检测、可靠性评估方向研究. E-mail:ljyskd@hotmail.com
基金资助:
国家自然科学基金项目(51404283)

Eddy current thermal imaging detection technology based on convolutional neural network

LI Jianyu^1,2,3, CHENG Jiaojiao^1,2, ZHU Heming^1,2, ZHENG Wenpei³

1. State Key Laboratory of Shale Oil and Gas Enrichment Mechanisms and Effective Development, Beijing 102206, China; 2. SINOPEC Research Institute of Petroleum Engineering Co., Ltd., Beijing 102206, China; 3. Key Laboratory of Oil and Gas Safety and Emergency Technology, Ministry of Emergency, Beijing 102249, China

Received:2023-11-01 Published:2025-11-19

摘要/Abstract

摘要： 针对涡流激励热成像检测在精度与速度方面存在的短板,对快速区域卷积神经网络算法(Faster R-CNN)进行了改进,将Inception-V4结构引入其中作为特征提取网络,提高缺陷识别的精度与速度.通过实验获取不同类型缺陷数据,导入卷积神经网络进行训练与测试,并与常见VGG和ResNet网络所处理的结果进行对比.研究结果表明,改进后的网络算法可以实现缺陷识别的需求,在识别精度和识别速度方面均优于传统算法.

关键词: 无损检测, 设备维护, 故障诊断, 缺陷识别

Abstract: For eddy current excitation thermal imaging detection, the faster region convolutional neural network(Faster R-CNN)algorithm is improved, and the Inception-V4 structure is introduced as a feature extraction network to improve the accuracy and speed of defect recognition. Experimental methods are used to obtain different types of defect data, and convolutional neural networks are imported for training and testing, and the results are compared with the results processed by common VGG and ResNet networks. The results show that the model proposed in this paper can meet the requirements of defect recognition, and is better than the traditional algorithm in terms of recognition accuracy and recognition speed.

Key words: non-destructive testing, equipment maintenance, fault diagnosis, defect identification

中图分类号:

TG115

李建宇,程姣姣,朱和明,郑文培. 基于卷积神经网络的涡流热成像检测技术[J]. 河北大学学报(自然科学版), 2025, 45(6): 653-660.

LI Jianyu, CHENG Jiaojiao, ZHU Heming, ZHENG Wenpei. Eddy current thermal imaging detection technology based on convolutional neural network[J]. Journal of Hebei University(Natural Science Edition), 2025, 45(6): 653-660.

参考文献

[1] 马国,贾华东,卢长煜,等.磁粉检测与渗透检测在工程机械结构件无损检测中的应用[J].无损检测, 2019, 41(2): 62-64. DOI:10.11973/wsjc201902014.
[2] CHENG L, TIAN G Y. Surface crack detection for carbon fiber reinforced plastic(CFRP)materials using pulsed eddy current thermography[J]. IEEE Sens J, 2011, 11(12): 3261-3268. DOI:10.1109/JSEN.2011.2157492.
[3] TIAN G Y, WILSON J, CHENG L, et al. Pulsed eddy current thermography and applications[M] //New Developments in Sensing Technology for Structural Health Monitoring. Berlin, Heidelberg: Springer Berlin Heidelberg, 2011: 205-231. DOI:10.1007/978-3-642-21099-0_10.
[4] 潘孟春,何赟泽,陈棣湘.涡流热成像检测技术[M].北京:国防工业出版社, 2013: 1-3.
[5] HE Y Z, LUO F L, PAN M C, et al. Defect classification based on rectangular pulsed eddy current sensor in different directions[J]. Sens Actuat A Phys, 2010, 157(1): 26-31. DOI:10.1016/j.sna.2009.11.012.
[6] 王晓娜,杨沛,侯德鑫,等.脉冲涡流热成像的自适应异常提取算法[J].仪器仪表学报, 2016, 37(8): 1818-1824. DOI:10.19650/j.cnki.cjsi.2016.08.012.
[7] D’ANGELO G, RAMPONE S. Shape-based defect classification for non destructive testing[C] //2015 IEEE Metrology for Aerospace(MetroAeroSpace), 2015, Benevento, Italy. IEEE, 2015: 406-410. DOI:10.1109/MetroAeroSpace.2015.7180691.
[8] WEI B, HAO K R, TANG X S, et al. A new method using the convolutional neural network with compressive sensing for fabric defect classification based on small sample sizes[J]. Text Res J, 2019, 89(17): 3539-3555. DOI:10.1177/0040517518813656.
[9] ABDEL-HAMID O, MOHAMED A R, JIANG H, et al. Applying Convolutional Neural Networks concepts to hybrid NN-HMM model for speech recognition[C] //2012 IEEE International Conference on Acoustics, Speech and Signal Processing(ICASSP), 2012, Kyoto, Japan. IEEE, 2012: 4277-4280. DOI:10.1109/ICASSP.2012.6288864.
[10] EHRHART B, ECAULT R, TOUCHARD F, et al. Development of a laser shock adhesion test for the assessment of weak adhesive bonded CFRP structures[J]. Int J Adhes Adhes, 2014, 52: 57-65. DOI:10.1016/j.ijadhadh.2014.04.002.
[11] AGRAWAL A, MITTAL N. Using CNN for facial expression recognition: A study of the effects of kernel size and number of filters on accuracy[J]. Vis Comput Int J Comput Graph, 2020, 36(2): 405-412. DOI:10.1007/s00371-019-01630-9.
[12] ACHARYA S, PANT A K, GYAWALI P K. Deep learning based large scale handwritten Devanagari character recognition[C] //2015 9th International Conference on Software, Knowledge, Information Management and Applications(SKIMA), 2015, Kathmandu, Nepal. IEEE, 2015: 1-6. DOI:10.1109/SKIMA.2015.7400041.
[13] ECKELS J D, JACOBSON E M, CUMMINGS I T, et al. Predicting local material thickness from steady-state ultrasonic wavefield measurements using a convolutional neural network[J]. Ultrasonics, 2022, 123: 106661. DOI:10.1016/j.ultras.2021.106661..
[14] RACHMADI R F, PURNAMA I E. Vehicle color recognition using convolutional neural network[EB/OL]. 2015: 1510.07391. https://arxiv.org/abs/1510.07391v3.
[15] PATEL D, HONG X P, ZHAO G Y. Selective deep features for micro-expression recognition[C] //2016 23rd International Conference on Pattern Recognition(ICPR), 2016, Cancun, Mexico. IEEE, 2016: 2258-2263. DOI:10.1109/ICPR.2016.7899972.
[16] QUEIROZ J C S, SANTOS Y T B, DA SILVA I C, et al. Damage detection in composite materials using tap test technique and neural networks[J]. J Nondestruct Eval, 2021, 40(1): 27. DOI:10.1007/s10921-021-00759-9.
[17] 胡德洲,左宪章,王建斌,等.脉冲涡流热成像检测激励参数的优化[J].无损检测, 2014, 36(8): 23-28.
[18] 刘俊岩,戴景民,王扬.红外图像序列处理的锁相热成像理论与试验[J].红外与激光工程, 2009, 38(2): 346-351. DOI:10.3969/j.issn.1007-2276.2009.02.034.
[19] HE M, LI J Y, ZHANG Y W, et al. Research on crack visualization method for dynamic detection of eddy current thermography[J]. NDT & E Int, 2020, 116: 102361. DOI:10.1016/j.ndteint.2020.102361. (

基于卷积神经网络的涡流热成像检测技术

Eddy current thermal imaging detection technology based on convolutional neural network

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 9

编辑推荐

Metrics

本文评价

[1]	郝紫阳,张晓宣,冯贺平. 基于滑动窗多时频脊线提取的滚动轴承变转速故障诊断[J]. 河北大学学报(自然科学版), 2025, 45(5): 551-560.
[2]	赵智强,帕孜来·马合木提,刘行行,周昂. 光伏逆变器IGBT器件接线故障诊断方法[J]. 河北大学学报(自然科学版), 2024, 44(1): 17-26.
[3]	庞彬,周紫烨,齐晓帆,郑涵升. 双馈异步风力发电机转子匝间短路故障振动特性分析[J]. 河北大学学报(自然科学版), 2024, 44(1): 9-16.
[4]	刘玉鑫,武文博,张雄,万书亭. 基于HHO-CNN的轴承故障诊断方法研究[J]. 河北大学学报(自然科学版), 2023, 43(6): 571-583.
[5]	郑一博,李忠灿,程杨鑫,董雨荷,朱占龙. 基于图像块鲁棒抑制式模糊C均值图像分割[J]. 河北大学学报(自然科学版), 2023, 43(4): 442-448.
[6]	佘丽萱,康佳,王楠,邵利敏. 一种新的棉花根系图像阈值分割方法[J]. 河北大学学报(自然科学版), 2022, 42(2): 124-130.
[7]	郝真鸣, 李承旺, 张超岩, 冉宁. 一种有界Petri网的可诊断性验证方法[J]. 河北大学学报(自然科学版), 2022, 42(1): 8-15.
[8]	杜二霞,张君鸿,阎宇杰. 基于超声波测试的混凝土应力与波速关系[J]. 河北大学学报(自然科学版), 2018, 38(4): 356-361.
[9]	石建国,商秀印,张瑞雪. 基于BP神经网络的沼气工程故障诊断专家系统[J]. 河北大学学报(自然科学版), 2013, 33(1): 102-106.