Kannusamy M, Kapuria Santosh, Sasmal Saptarshi
CSIR-Structural Engineering Research Centre, Taramani, Chennai 600113, India; Academy of Scientific and Innovative Research (AcSIR), Ghaziabad 201002, India.
Department of Applied Mechanics, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110016, India.
Ultrasonics. 2022 Aug;124:106767. doi: 10.1016/j.ultras.2022.106767. Epub 2022 May 23.
This study proposes a modified virtual time-reversal (VTR) algorithm for baseline signal-free damage detection in plate-like structures. The physical actuation and sensing of Lamb waves are performed using a broadband Gaussian excitation instead of the conventional narrowband modulated tone burst excitations. The forward response and the reconstructed signal due to the time-reversal process for a narrowband input signal are then constructed virtually using the broadband transfer function. The method eliminates the possibility of numerical errors encountered in the conventional VTR method based on narrowband excitations. It is also more efficient than the traditional VTR algorithm because it can probe at multiple excitation frequencies using a single measurement for each sensing path. This modified VTR algorithm is employed in the recently developed refined time-reversal method (RTRM), which uses an extended signal length of the reconstructed signal for computing damage index (DI) and probes the structure at the best reconstruction frequency. The new method is termed the virtual refined time-reversal method (VRTRM). The DIs based on the VRTRM are used in the reconstruction algorithm for probabilistic inspection of defects to achieve baseline signal-free localization of damages. The efficacy of the proposed VRTRM for damage localization is experimentally verified with the established technique RTRM. Experiments are performed in an aluminium plate equipped with a network of surface-bonded piezoelectric patch transducers to illustrate the conventional VTR's pitfalls and the modified VTR's accuracy for a single mass damage scenario. The results show that the proposed VRTRM is as accurate as the established technique RTRM in estimating the reconstructed signals and localizing a block mass damage. Finally, the VRTRM is shown to localize in a dual damage scenario with excellent accuracy. In contrast, the conventional main mode-based VTR method fails to localize the damages with or without single-mode tuning.
本研究提出了一种改进的虚拟时间反转（VTR）算法，用于板状结构中无基线信号的损伤检测。使用宽带高斯激励而非传统的窄带调制脉冲串激励来实现兰姆波的物理激励和传感。然后，利用宽带传递函数虚拟构建窄带输入信号经时间反转过程后的正向响应和重构信号。该方法消除了基于窄带激励的传统VTR方法中出现数值误差的可能性。它也比传统的VTR算法更高效，因为它可以在每个传感路径单次测量的情况下，在多个激励频率下进行探测。这种改进的VTR算法应用于最近开发的改进时间反转方法（RTRM）中，该方法使用重构信号的扩展信号长度来计算损伤指数（DI），并在最佳重构频率下探测结构。新方法被称为虚拟改进时间反转方法（VRTRM）。基于VRTRM的DI用于缺陷概率检测的重构算法中，以实现损伤的无基线信号定位。通过已有的技术RTRM对所提出的VRTRM进行损伤定位的有效性进行了实验验证。在配备有表面粘贴压电贴片换能器网络的铝板上进行了实验，以说明传统VTR的缺陷以及改进VTR在单一质量损伤情况下的准确性。结果表明，所提出的VRTRM在估计重构信号和定位块状质量损伤方面与已有的技术RTRM一样准确。最后，VRTRM在双损伤情况下也能以极高的精度进行定位。相比之下，传统的基于主模态的VTR方法无论是否进行单模态调谐都无法定位损伤。
