基于时间反演的无损检测成像初探  

Exploration of Microwave Non-Destructive Evaluation Imaging Based on Time Reversal

作　　者：陈恬田 杨晓庆[1] 田姗 李旭东 CHEN Tiantian;YANG Xiaoqing;TIAN Shan;LI Xudong(College of Electronics And Information Engineering of Sichuan University,Chengdu 610065,China)

机构地区：[1]四川大学电子信息学院,四川成都610065

出　　处：《传感器世界》2023年第5期25-30,共6页Sensor World

摘　　要：时间反演理论自上世纪90年代提出就引起了研究人员的兴趣,于2004年由声学拓展到电磁学,并被应用于无损检测、通信和医疗成像等多个领域。文中微波的时间反演多信号分类法被首次用于介质材料的无损检测,通过测量电磁波分别在有损样本和标准无损样本传播的信道响应,并取其差值获得用信道响应矩阵表示的样本缺陷对电磁波的散射作用,进而利用特征值分解对信道响应进行降维,并根据已知信道格林函数进行成像。根据在时域有限积分软件的仿真结果可知,该方法能够分辨样品中尺寸远小于波长的单个微小缺陷,且误差较小,是一种超分辨率的方法。Theory of time reversal was proposed in 1990s,it has aroused interest of researchers and was extended from acoustics to electromagnetism in 2004,being applied in many fields such as nondestructive testing,communication and medical imaging.In the paper,the microwave time reversal multi-signal classification method is used for the first time in the nondestructive testing of dielectric materials.After measuring the channel response of electromagnetic wave propagation in the lossy sample and the standard lossless sample respectively,the difference is calculated,thus the electromagnetic scattering effect of defect is obtained represented by the channel response matrix.By eigenvalue decomposition of channel response,imaging is achieved with the known Green’s Function.According to the simulation in software of time domain finite integration,the super-resolution method is able to distinguish a single tiny defect whose size is much smaller than the wavelength with a small error.

关 键 词：无损检测 微波时间反演 时间反演多信号分类算法 

分 类 号：TN98[电子电信—信息与通信工程] O441.5[理学—电磁学]