[2A1] Ultrasonic imaging of volumetric defect using diffuse signals acquired from a laser vibrometer
J Li
University of Bristol, UK
Laser-induced ultrasonic arrays have been used for defect detection and have advantages of being non-contact and able to cope with complex geometry and harsh environments. However, the detection speed is limited by slow data acquisition, as the combinations of every transmitting and receiving sampling point are required to build up full matrix capture (FMC) array data. This issue becomes even more challenging when detecting volumetric defects using a two-dimensional array. In the proposed inspection configuration, a single conventional piezoelectric transducer is used to transmit ultrasound into a specimen at an optimised location, while a laser interferometer collects data at predetermined 2D sampling layouts. Chirp signals are employed as an excitation method to suppress noise in the acquired signals, effectively eliminating the need for extensive signal averaging and improving data acquisition efficiency. An FMC array dataset is generated by cross-correlating pairs of acquired signals in their wave diffuse regimes, significantly reducing the number of data acquisition measurements by a factor equivalent to the number of sampling points. This approach also enhances the directivity function of the equivalent transmitting array elements compared to laser-induced phased arrays (LIPAs). A 3D total focusing method (TFM) image is then constructed using the generated FMC dataset. Notably, in the constructed FMC dataset, the signals are solely influenced by the directivity function of the receiving laser, rather than the transmitting laser, as seen in LIPAs.
The performance of the proposed inspection method was demonstrated in detecting various defects in aluminium specimens. It is shown that the proposed method can efficiently generate FMC array datasets with a large number of equivalent array elements that is beyond the limit of current piezoelectric ultrasonic arrays and can effectively detect volumetric defects from the resulting 3D TFM images.
The performance of the proposed inspection method was demonstrated in detecting various defects in aluminium specimens. It is shown that the proposed method can efficiently generate FMC array datasets with a large number of equivalent array elements that is beyond the limit of current piezoelectric ultrasonic arrays and can effectively detect volumetric defects from the resulting 3D TFM images.
