基于多模态信号的金属材料缺陷无损检测方法

金属材料缺陷检测对于经济发展具有重要意义.针对现有无损检测技术信号模态单一、检测范围有限等不足,提出了一种基于多模态信号的金属材料缺陷无损检测方法.该方法以光声无损检测方法为主体,首先利用有限元方法分析了缺陷对激光能量吸收量和光声表面波传播的影响,提出了基于激光吸收量和光声表面波的缺陷检测方法;然后搭建了多模态信号检测平台,采集了缺陷的光学、光声和超声三种模态的信号,检测出了裂纹的宽度和分布信息,以及深度和在内部的延伸状况.研究结果表明,本文提出的基于金属材料多模态信号的无损检测方法能够准确、全面地检测出金属材料的杂质和裂纹的尺寸信息,弥补了现有无损检测方法在检测范围上的不足,为缺陷定量检测和全面诊断提供了一个新的思路.

Nondestructive detecting method for metal material defects based on multimodal signals

Metal materials play an important role in many domains, which are significant to the national economy. However, different kinds of metal defects, such as cracks, contraction cavities, impurities, will be generated in the process of production and service. These defects will affect the metal service life and mechanical properties directly, and even cause serious economic loss. Therefore, it is vital to detect the metal defects. Numerous nondestructive testing (NDT) methods have been proposed for detecting metal defects, such as ultrasonic (US) testing, eddy current testing, photoacoustic (PA) testing, magnetic particle testing, etc. However, each of them uses a single modal signal, which leads to a limited detection range. A nondestructive detecting method for metal material defects based on multimodal signals is proposed to expand the scope of detection and obtain more complete information. Specifically, optical signal, PA signal and US signal are combined together in this method, with the consideration of their complementarities. Simulation and experiments are conducted to validate the effectiveness of the proposed method. Firstly, finite element simulation is employed to analyze the relationship between material parameters and the absorption of laser energy. Meanwhile, the influence of defect size on PA surface wave is simulated and analyzed. Then, a multimodal NDT platform is established to collect and process optical, PA and US signals of the metal defects. These three modal signals contain information about metal surface, shallow surface and internal defects respectively. Eventually, the information, including the location, appearance on the surface, depth, extension path in the material, is obtained. As demonstrated in the results, the nondestructive detecting method based on multimodal signals can detect the metal defects accurately and comprehensively. This method improves the existing methods in terms of detection range and quantitative detection. Additionally, it provides a new way for the quantitative detection and comprehensive diagnosis of metal defects.