A curvature-adaptive phased-array synthetic aperture focusing method for flaw detection and size quantification in cylinder-like parts

A phased-array ultrasonic synthetic aperture focusing technique (SAFT) method is proposed to improve the lateral spatial resolution in detecting flaws in parts with curved surfaces. The overall method consists of an adaptive grid size optimization for SAFT reconstruction, and a distance-gain-size method for flaw quantification. Combining the two ingredients, the method allows for mitigating the under- and over-sampling effects arising from the mismatch of Cartesian reconstruction grids and testing data acquired from a curved surface, and enhancing the lateral spatial resolution and size quantification accuracy. Ultrasonic testing is performed using a 10 MHz linear phased-array probe and a wire and arc additively manufactured aluminum disk with six closely spaced artificial flaws. The results show that the method can reliably identify six flat-bottomed holes with nominal sizes ranging from 0.4 mm to 2 mm and a minimal edge-to-edge distance of 0.8 mm. Comparisons are made with regular SAFT methods and conventional phased-array ultrasonic testing to demonstrate the advantage of the proposed method. The reliability of the proposed method is further examined using the model-assisted probability of detection. The results show that a 0.51 mm flaw can be detected by the method with a 90% probability of detection at 95% confidence.