基于超声相控阵的角焊缝缺陷信号重构方法研究

摘要:

在对接管角焊缝进行超声检测的过程中，由于检测的工件结构复杂并且仪器设备本身会受到电信号干扰，所以检测得到的A扫信号存在噪声，在检测图像中会出现“伪像”，从而造成检测困难。如果想要提高检测图像的质量，对于A扫信号的除噪尤为重要，所以提出小波去噪和经验模态分解相结合的角焊缝缺陷信号重构方法。首先分析接管角焊缝结构特点，进行相控阵检测试验，得到缺陷检测数据；然后着重分析了裂纹与未熔合缺陷信号，完成对其的经验模态分解与重构；最后通过对原始信号进行小波去噪及经验模态分解与重构，从而达到了比传统的算法更高的信噪比与更低的均方误差，说明本算法有更好的去噪效果，更有利于超声回波信号的分析。

Abstract:

In the process of ultrasonic detection of nozzle fillet welds, due to the complex structure of the detected workpiece and the interference of the instrument itself by electrical signals, the detected A-scan signal has noise, and there will be "artifacts" in the detection image, thus causing detection difficulties. If we want to improve the quality of detection images, denoising of A-scan signals is particularly important. Therefore, this paper proposes a fillet weld defect signal reconstruction method combining wavelet denoising and empirical mode decomposition. Firstly, the structural characteristics of fillet weld are analyzed, and the phased array test is carried out to obtain the defect detection data. Secondly, the empirical mode decomposition and reconstruction of cracks and unfused defects are analyzed. Finally, the wavelet denoising and empirical mode decomposition and reconstruction of the original signal are carried out to achieve higher signal-to-noise ratio and lower mean square error than the traditional algorithm, which shows that this algorithm has better denoising effect and is more beneficial to the analysis of ultrasonic echo signal.