A Neural Network Algorithm for Ultrasonic Imaging of Pile Defects
-
摘要: 超声波法以无损性、准确性高等特点在桩基质量检测中应用广泛。传统超声波透测法依据声速、波幅及PSD等二维波形图进行缺陷判断,检测结果的可读性、直观性相对较差。基于修正的残差卷积神经网络算法(M-ResNet),通过射线追踪法构建原始超声波数据集,利用深度学习技术训练网络模型,采用MATLAB编制代码显示3D桩基础缺陷图像和检测结果。缺陷图像评价指标ICC、CCC及算法抗噪测试表明:该算法速度快、检测精度高,能直观显示缺陷范围和大小,在成像方面比传统方法更具优势。该方法可直接利用传统超声波法的仪器和测试数据进行3D图形显示和缺陷检测,研究成果可在类似工程检测中推广应用。Abstract: Ultrasonic testing has been widely used in the quality inspection of pile foundations due to its non-destructiveness and high accuracy. Traditional ultrasonic transmission method relies on two-dimensional waveform diagrams (e.g., sound velocity, amplitude, and PSD) for defect judgment, and the readability and intuitiveness of the detection results are relatively poor. Based on a modified residual convolutional neural network algorithm (M-ResNet), an original ultrasonic dataset using the ray-tracing method was constructed. Deep learning technology was employed to train the network model, and MATLAB was used to develop code for displaying 3D pile foundation defect images and detection results. Evaluation indicators (ICC and CCC) for defect images and algorithm noise immunity tests demonstrated that the algorithm operated rapidly with high detection accuracy, could directly display the defect range and size, and showed more advantages in imaging compared to traditional methods. The proposed method can directly use existing instruments and test data from conventional ultrasonic methods to achieve 3D visualization and defect detection, with research results applicable to similar engineering inspections.
-
Key words:
- civil engineering /
- pile foundation inspection /
- deep learning /
- neural network
-
[1] DAVIS A G,ALLEN G. Nondestructive evaluation of existing deep foundations[J]. Journal of Performance of Constructed Facilities,1995,9(1):57-74. [2] 吴君涛,王奎华,肖偲,等. 弹性支承桩周围土振动响应解析解及其波动规律研究[J]. 岩石力学与工程学报,2018,37(10):2384-2393. [3] 王凤武,刘学冬. 无损检测技术在桥梁桩基检测中的运用分析[J]. 工程建设与设计,2024(2):161-163. [4] 吴君涛,王奎华,高柳,等. 考虑桩身材料阻尼的桩基纵向振动积分变换解及其应用[J]. 岩石力学与工程学报,2017,36(9):2305-2312. [5] WU J T,WANG K H,GAO L,et al. Study on longitudinal vibration of a pile with variable sectional acoustic impedance by integral transformation[J]. Acta Geotechnica,2019,14(6):1857-1870. [6] 梁道盛. 工程基桩检测方法对比[J]. 公路交通科技(应用技术版),2013,9(3):419-423. [7] 王奎华,于喆,项驰轩,等. 一发三收式桩基声波透射检测技术及数值模拟分析[J]. 中国测试,2023,49(1):1-6. [8] 黄亮. 基桩声波透射法CT技术的桩身缺陷识别研究[D]. 武汉:华中科技大学,2019. [9] 王珂,吴君涛,于喆,等. 混凝土基桩超声波层析成像技术优化[J]. 浙江大学学报(工学版),2023,57(4):805-813. [10] 王冠,陈尧,王志刚,等. 混凝土桩基SAGA层析成像算法实验研究[J]. 传感技术学报,2020,33(4):517-523. [11] 于喆,王奎华,吴君涛. 基于改进克里金法的基桩缺陷空间插值研究[J]. 地基处理,2023,5(5):376-382,450. [12] 叶启扬,吴君涛,易恩泽,等. 基于三维空间路径追踪的混凝土基桩层析成像技术研究[J/OL]. 土木工程学报,2025[ 2025-07-07]. https://doi.org/10.15951/j.tmgcxb.24070537. [13] 刘珂铖,谢群,李雁军. 基于深度学习YOLOX算法的混凝土构件裂缝智能化检测方法[J]. 济南大学学报(自然科学版),2024,38(3):341-349. [14] 于鸣,向广德,于金生. 基于深度学习的道路损坏检测算法研究[J/OL]. 哈尔滨理工大学学报,2025[ 2025-07-07]. http://kns.cnki.net/kcms/detail/23.1404.N.20240407.1334.023.html. [15] 仲伟峰,弓越,刘杰. 改进YOLOv5算法的卡通动漫人物脸部检测方法[J/OL]. 哈尔滨理工大学学报,2025[ 2025-07-07]. http://kns.cnki.net/kcms/detail/23.1404.N.20240621.1343.010.html. [16] ZHAO W,WANG H,GEMMEKE H,et al. Ultrasound transmission tomography image reconstruction with a fully convolutional neural network[J]. Physics in Medicine& Biology,2020,65(23),235021. [17] 尹宝才,王文通,王立春. 深度学习研究综述[J]. 北京工业大学学报,2015,41(1):48-59. [18] HUANG J T,LI J,GONG Y. An analysis of convolutional neural networks for speech recognition[C]// IEEE International Conference on Acoustics. 2015. [19] 刘聪然. 混凝土超声波层析成像正反演方法数值研究[D]. 武汉:武汉理工大学,2016. [20] MING J,GE W. Convergence of the simultaneous algebraic reconstruction technique(SART)[J]. IEEE Transactions on Image Processing A Publication of the IEEE Signal Processing Society,2001,1(8),957961. [21] BOWMAN A W,AZZALINI A. Applied smoothing techniques for data analysis[M]. New York:Oxford University Press Incorporated,1997. -
点击查看大图
计量
- 文章访问数: 79
- HTML全文浏览量: 15
- PDF下载量: 3
- 被引次数: 0
登录
注册
E-alert
登录
注册
E-alert
下载: