水利学报2026，Vol.57Issue(2)：255-265,279,12.DOI:10.3724/j.slxb.20250429
基于相干加权融合的混凝土缺陷检测全聚焦成像技术研究
Research on total focusing method for concrete defect detection based on coherence weighted fusion
摘要
Abstract
Concrete is a composite material composed of multi-phase heterogeneous components such as aggregates and cement paste.While the complex acoustic impedance interfaces within it provide the physical basis of ultrasonic defect detection,they also generate strong scattering noise and significant ultrasonic energy attenuation,imposing considerable limitations on the imaging quality of ultrasound-based detection.Although the total focusing method(TFM)technology can improve the resolution through full matrix capture(FMC)data acquisition,but it remains sus-ceptible to noise and artifacts in strongly scattering media such as concrete.In order to further enhance TFM quality,a TFM algorithm based on coherence weighted fusion(CWF)is proposed.This method aims to maximize the use of the scattered field contained in the FMC data to amplify the difference between coherent and incoherent signals,thereby enhancing the final imaging resolution and defect identifiability.Numerical simulations on a meso-scale con-crete model and experimental results show that the proposed CWF-TFM optimization algorithm can effectively sup-press certain noise and artifacts,improve the longitudinal resolution,and concentrate signals in defect areas.When using a 50 kHz excitation signal to detect a single crack defect,the signal-to-noise ratio(SNR)is improved by 30.0%compared with traditional TFM.For single void defect detection,the SNR is increased by 32.2%over tradi-tional TFM.and for multiple defects,the SNR shows a 22.7%enhancement.This achievement overcomes the modal limitations of traditional TFM in concrete defect detection,significantly improving image SNR and defect identifiabil-ity,and provides a new approach for high-resolution ultrasonic detection of concrete defects.关键词
全聚焦成像/混凝土缺陷/相干加权融合/成像分辨率/信噪比Key words
total focusing method/concrete defects/coherent weighted fusion/imaging resolution/signal-to-noise ratio分类
建筑与水利引用本文复制引用
黄昊,刘金龙,孙林远,张俊俊,王振彬,杨伟才..基于相干加权融合的混凝土缺陷检测全聚焦成像技术研究[J].水利学报,2026,57(2):255-265,279,12.基金项目
中国水科院基本科研业务费项目(SM0145B042021,SM110145B0012025) （SM0145B042021,SM110145B0012025）
流域水循环模拟与调控国家重点实验室自主研究课题(SKL2024YJTS14) （SKL2024YJTS14）
