Application Cases

Experiment Name: Monitoring Damage Progression in Welded Seams with Defects Based on Coda Wave Interferometry

Experimental Content:
To monitor the development of defect damage in welded seams, it is necessary to emit ultrasonic waves using piezoelectric ceramics and receive them at the opposite end of the weld seam. Under cyclic loading of the welded structure, defects undergo changes, thereby altering the propagation paths of the multiply scattered ultrasonic waves. This study applies the coda wave interferometry theory to monitor weld seam damage and uses the waveform stretching method to quantify changes in the coda wave. Based on the variation of the average stretching coefficient index, the progression of weld seam defects under cyclic loading can be monitored. The schematic diagram of weld seam monitoring is shown in the figure below. Piezoelectric ceramic PZT1 emits ultrasonic waves, while piezoelectric ceramic PZT2 receives them. Before being received, the ultrasonic waves undergo multiple scattering within the specimen, forming interference waves. By analyzing the coda wave portion of the interference waves, the damage condition of defects in the weld seam can be determined.

Testing System:
The experimental equipment includes an MTS fatigue loading machine, an integrated chassis equipped with an oscilloscope and an arbitrary waveform generator, a high-voltage power amplifier (ATA-2021H), and a computer running LabVIEW data acquisition software.

Conclusion:
This study proposes a method based on piezoelectric sensor coda wave interferometry to monitor minute changes in damage caused by incomplete penetration defects under varying cyclic loading conditions. First, the results of cyclic loading experiments show that as the number of cycles increases, the axial displacement of the MTS machine gradually increases under the same loading force, indicating the emergence and gradual growth of minor plastic deformation in the specimen. Next, an analysis of the time-domain received signals from coda wave interferometry experiments was conducted, comparing the overall waveforms, direct wave waveforms, and coda wave waveforms of each specimen under different numbers of cycles. The results reveal that, compared to the direct wave, the coda wave signals exhibit significant time shifts across different cycles. Subsequently, the waveform stretching method was used to further calculate the time shifts between different coda waves, obtaining the cross-correlation coefficients for each specimen under different cyclic loading conditions, as well as the stretching coefficient ε_max corresponding to the maximum cross-correlation coefficient, which represents the relative velocity change of the coda wave. Based on three repeated experimental results, the variation trend of the average stretching coefficient |ε_max| for each specimen under different numbers of cycles was obtained. It was found that as the number of cycles increases, the average stretching coefficient |ε_max|, which represents the relative velocity change of the coda wave, gradually increases, showing a consistent overall trend. According to the correspondence between the number of cycles and the average stretching coefficient |ε_max|, it can be observed that by calculating the relative velocity change of the coda wave, the minute changes caused by incomplete penetration defects in the specimen at different times can be effectively reflected, enabling quantitative analysis of the damage degree of the specimen.

Figure: Average Values Under Different Stretching Coefficients

Figure: ATA-2021B High-Voltage Amplifier Specifications and Parameters