Empa, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland; Institute for Mechanical Systems, ETH Zürich, 8092 Zürich, Switzerland.
Empa, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland.
Ultrasonics. 2020 Jan;100:105984. doi: 10.1016/j.ultras.2019.105984. Epub 2019 Aug 20.
In this paper we demonstrate a non-destructive, non-contact detection method for small defects in thin polymer plates using an air coupled ultrasonic (ACUS) setup. There exist many applications for such methods, e.g. quality control in the manufacturing process or failure prevention by periodical inspections during the lifetime of a product. We demonstrate a setup for the inspection of plates together with signal analysis algorithms to process the measured data, meeting the challenges to handle the dispersive signals and establishing a robust failure criterion. Pressure waves from the transmitter excite different modes of Lamb waves inside the plate. These Lamb waves propagate in the plate and reradiate pressure waves into the air that are then detected by the receiver. Lamb mode conversion is used for defect detection. A numerical model allows the visualization of the propagating waves in the air as well as the Lamb waves inside the plate. Four key parameters of the setup are identified, two angles and two distances. The transmitter and the receiver angles are used to select which Lamb mode (anti-symmetric A or symmetric S) is mainly excited and detected, respectively. For the acquisition of the Lamb wave signal the distance from the transmitter to the receiver should be as large as possible but is limited by the attenuation of the signal. Measurements for different values of this distance are essential for signal analysis. The distance between transducer and plate surface should be as small as possible even if it may introduce secondary Lamb waves due to reflections of the pressure wave between transmitter and plate surface. Two algorithms, a model based one and a data driven one, are presented to separate Lamb modes that overlap in time. In these separated signals, the Lamb mode conversion from A to S is shown, allowing a localization of the defect. We conclude that defect detection and localization with Lamb mode conversion is possible with an air coupled ultrasonic setup. Multiple measurements along the propagation direction of the Lamb waves are necessary to allow a thorough signal analysis and visualize the mode conversion.
在本文中,我们展示了一种使用空气耦合超声(ACUS)装置对薄聚合物板中的小缺陷进行非破坏性、非接触检测的方法。这种方法有许多应用,例如在制造过程中的质量控制或在产品寿命期间通过定期检查来防止故障。我们展示了一种用于检查板的设置以及信号分析算法,以处理测量数据,满足处理分散信号和建立稳健故障标准的挑战。发射器产生的压力波在板内激发不同模式的兰姆波。这些兰姆波在板内传播,并将压力波重新辐射到空气中,然后由接收器检测到。兰姆波模式转换用于缺陷检测。数值模型允许可视化在空气中传播的波以及板内的兰姆波。确定了设置的四个关键参数,两个角度和两个距离。发射器和接收器的角度用于选择主要激发和检测的兰姆模式(反对称 A 或对称 S)。对于兰姆波信号的采集,从发射器到接收器的距离应尽可能大,但受到信号衰减的限制。对于这个距离的不同值的测量对于信号分析是必不可少的。换能器和板表面之间的距离应尽可能小,即使由于发射器和板表面之间的压力波反射会引入二次兰姆波。提出了两种算法,一种基于模型的算法和一种数据驱动的算法,用于分离时间上重叠的兰姆波模式。在这些分离的信号中,显示了从 A 到 S 的兰姆波模式转换,允许对缺陷进行定位。我们得出结论,使用空气耦合超声装置可以实现兰姆波模式转换的缺陷检测和定位。沿兰姆波传播方向进行多次测量是必要的,以允许进行彻底的信号分析并可视化模式转换。