Edge blurring reduction in contact-state ultrasonic measurement via matching pursuit

Ultrasonic measurement provides an effective and accurate method for the performance evaluation of mechanical devices, but the edge blurring effect generated in the inspection process will lead to measurement errors and reduce the reliability. In this study, the edge blurring effect is studied, and a matching pursuit model is proposed and established, through which the interference signal is successfully predicted and separated, thus effectively suppressing the influence of the blurred signal on the detection results. Based on COMSOL Multiphysics, a two-dimensional finite element model is established to simulate the reflected signals of ultrasonic waves at the ideal contact interface between structural steel and aluminum. The matching accuracy of the MP algorithm is verified, and the characteristics of the ultrasonic echo signals are also characterized by an atomic function model to predict the reflection coefficients. The experimental results demonstrate that the proposed method reduces the maximum error of the reflection coefficient by nearly 200%, providing an effective approach to suppress edge blurring in the field of ultrasonic inspection. This is of great significance for the evaluation of product quality.