The characterisation of surface-breaking crack using ultrasonic total focusing method imaging based on COA-VMD

The use of ultrasonic total focusing method (TFM) to characterise surface-breaking crack plays a significant role in non-destructive evaluation. It not only provides early detection capability but also offers critical information for the assessment of component integrity. This paper proposes an improved ultrasonic TFM based on COA-VMD denoising for accurate defect quantification. The proposed method decomposes the full matrix capture (FMC) data acquired by the ultrasonic array transducer using variational mode decomposition (VMD), with the VMD parameters optimised by the Coati Optimisation Algorithm (COA). The effective intrinsic mode function components obtained from VMD are selected based on the evaluation parameters and then superimposed to reconstruct the FMC data. The reconstructed FMC data is subsequently used for TFM imaging. To improve image quality and defect quantification, the Butterworth filter is applied to enhance the features of the defect. Experimental results show that the proposed method can accurately quantify the position and width of surface-breaking crack, with position quantification errors less than 4.38% and width quantification errors less than 4.12%.