Model-based compressive sensing for damage localization in Lamb wave inspection.

Area of Science:

• Non-destructive testing
• Signal processing
• Materials science

Background:

• Compressive sensing (CS) offers efficient signal compression and analysis for sparse signals.
• Lamb-wave-based methods are crucial for defect detection in structures.

Purpose of the Study:

• To develop a compressive sensing (CS) approach for ultrasonic signal decomposition.
• To enhance performance in Lamb-wave-based defect detection procedures.

Main Methods:

• Utilized a CS algorithm based on alternating minimization (AM) for signal decomposition.
• Employed the warped frequency transform to generate a sparsifying basis, compensating for dispersion.
• Extracted system impulse response and reflectivity function information.

Main Results:

• Demonstrated effective signal decomposition on synthetic and experimental Lamb wave data.
• Achieved improved accuracy in wave propagation path length estimation compared to existing strategies.
• Validated the approach on Lamb waves propagating in an aluminum plate.

Conclusions:

• The proposed CS approach significantly improves ultrasonic signal decomposition for Lamb-wave-based defect detection.
• Enhanced accuracy in wave propagation analysis facilitates more precise defect localization.
• This technique shows promise for advancing non-destructive evaluation methods.