基于概率损伤算法的复合材料板空气耦合Lamb波扫描成像

Scanning imaging of composite plate using air-coupled Lamb waves based on probabilistic damage algorithm

  • 摘要: 采用非接触空耦传感器在准各向同性复合材料板中激励出单一的Lamb波模态, 用于分层缺陷的扫描检测。扫描时, 激励和接收传感器置于复合材料板同侧并相对倾斜布置, 传感器沿2个正交方向同步线性扫描, 得到不同位置的检测信号。对不同扫描路径下的检测信号进行连续小波变换, 提取激励频率下的小波系数包络信号, 对分层缺陷进行成像。在此基础上, 利用概率损伤算法定义损伤指数, 结合不同方向的损伤指数实现分层缺陷成像。采用全加法和全乘法对2个正交扫描方向得到的成像结果进行数据融合, 实现了分层缺陷的定位和重构。并在成像算法中引入阈值, 进一步提高了分层缺陷的定位精度以及重构质量。

     

    Abstract: Non-contact air-coupled ultrasonic transducers were used to generate pure Lamb wave mode in quasi-isotropic composite plate, the wave mode was used to scan for delamination inspection. In the scanning process, generating and receiving transducers were arranged on the same side of the plate oppositely and obliquely. They were linearly moved synchronously in two orthogonal directions in order to obtain different inspection signals at different positions. Wavelet coefficient envelope signals at certain generating frequency were extracted from the inspection signals in different scanning paths through continuous wavelet transform to image the delamination defects. Based on this, damage index was defined by using probabilistic damage algorithm and damage index of different directions were combined to realize the delamination imaging. Data fusion of imaging results acquired from two orthogonal directions was achieved by using full summation algorithm and full multiplication algorithm. The delamination defect can be well located and reconstructed. Furthermore, a threshold was introduced to the imaging algorithm to improve the location precision and reconstruction quality of the delamination defect.

     

/

返回文章
返回