基于改进自适应阈值EIT算法的CFRP损伤检测  

Research on CFRP Damage Detection Based on Improved Adaptive Threshold EIT Algorithm

作　　者：马敏 山雨泽 MA Min;SHAN Yuze(College of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China)

机构地区：[1]中国民航大学电子信息与自动化学院,天津300300

出　　处：《计量学报》2024年第5期730-737,共8页Acta Metrologica Sinica

基　　金：国家自然科学基金(61871379);天津市教委科研计划(2020KJ012)。

摘　　要：电阻抗层析成像(EIT)具有快速、无辐射等众多优点,但EIT的逆问题严重的病态性导致FISTA等算法的重建图像中存在损伤边缘信息缺失的现象。针对该问题引入了一种与解向量稀疏度相关的自适应阈值算子和一种可变阈值函数,解决了软阈值函数边缘处不可导的问题。仿真实验表明改进算法与传统的算法相比,FIMSTA算法的表现最好,尤其是对于传统算法图像重建效果较差的中心裂纹损伤,FIMSTA算法的相关系数达到0.7038,较表现最好的FISTA算法提升了51.08%。Electrical impedance tomography(EIT)has many advantages such as fast speed and no radiation,and has gradually emerged in the research of carbon fiber reinforced composite materials in recent years.The serious pathological nature of the inverse problem of EIT leads to the phenomenon of missing damaged edge information in the reconstructed images of algorithms such as FISTA.An adaptive threshold operator related to the sparsity of the solution vector and a variable threshold function have been introduced to address the above issues,solving problem of non differentiability at the edges of the soft threshold function.The simulation experiment shows that compared with traditional algorithms,the improved algorithm performs the best overall.For the center crack damage with poor image reconstruction performance of traditional algorithms,the correlation coefficient of the FIMSTA algorithm reaches 0.7038,which is 51.08%higher than the best performing FISTA algorithm.

关 键 词：无线电计量 电阻抗层析成像 CFRP 损伤检测 稀疏正则化算法 图像重建 自适应阈值算子 可变阈值函数 

分 类 号：TB973[一般工业技术—计量学]