基于组稀疏的桥梁混凝土波速反演重建方法
2025,33(1):261-268
摘要：无损检测的开展能够在不破坏混凝土结构的基础上对其内部的病害缺陷进行测定，是当前桥梁混凝土检测中常见的应用手段；由于混凝土内部结构复杂、测点布置受限以及检测区域大，为了提高缺陷检测精度，采用了联合代数重建算法结合组稀疏正则化（SART-GSR）的方法来实现稀疏测点条件下桥梁混凝土层析成像，结合桥梁混凝土层析成像原理建立数学模型，利用SART算法对其速度值进行求解，在SART结果的基础上，使用GSR对其进行优化解算处理。经过仿真实验验证，将SART-GSR算法与SART算法以及ART算法的重建效果进行对比，结果表明，SART-GSR算法相较于SART算法以及ART算法能够提升桥梁混凝土层析成像精度，对桥梁混凝土缺陷检测具有一定的应用参考价值。
关键词：无损检测;走时层析成像;桥梁混凝土;联合迭代重建算法;组稀疏;字典学习;
Reconstruction Method Of Bridge Concrete Wave Velocity Inversion Based On Group Sparsity
Abstract：Non-destructive testing can determine the internal defects of concrete structures without causing damage, and it is a common application in bridge concrete inspections. Due to the complexity of the internal structure of concrete, limitations in the layout of measurement points, and large inspection areas, to improve the accuracy of defect detection, a combined algebraic reconstruction technique algorithm with group sparse regularization (SART-GSR) has been adopted to achieve tomographic imaging of bridge concrete under sparse measurement points. A mathematical model is established based on the principle of bridge concrete tomography. The SART algorithm is used to solve for its velocity values, and on top of the SART results, GSR is applied for optimization and calculation processing. After simulation experiments, the SART-GSR algorithm was compared with the SART algorithm and the ART algorithm in terms of reconstruction effects. The results show that the SART-GSR algorithm can enhance the tomographic imaging precision of bridge concrete compared to the SART and ART algorithms, providing valuable reference for the detection of bridge concrete defects.
Key words：Non-destructive testing; Travel time tomography; Bridge concrete; Joint iterative reconstruction algorithm; Group sparsity; Dictionary learning
收稿日期：2024-08-26
基金项目：2024年山西省专利转化计划项目（200405004）；国家自然基金面上科学基金（62271453）；中央支持地方项目（YDZJSX2024D031）；山西省青年学术带头人项目（2024Q022）
