Experimental study of passive defect localization in plates using ambient noise

Area of Science:

• Materials Science
• Mechanical Engineering
• Acoustics

Background:

• Passive listening is a practical method for structural health monitoring.
• Detecting defects in structures is crucial for safety and maintenance.

Purpose of the Study:

• To experimentally apply passive listening methodology for defect detection in thin aluminum plates.
• To localize defects using correlation matrices and beamforming algorithms.

Main Methods:

• Estimating correlation matrices from recorded noise vibrations on a transducer array.
• Applying a beamforming algorithm to the difference between correlation matrices (with and without defect).
• Utilizing a vibrometer to assess the fidelity of transient responses.

Main Results:

• Successfully detected defects using different noise sources.
• Demonstrated robustness in detecting massive inclusions, holes, and cracks.
• Showed that defect localization is successful even with non-uniform, spatially stationary noise sources.

Conclusions:

• Passive listening with correlation matrices and beamforming is an effective technique for structural health monitoring.
• The method is robust to various defect types and noise source distributions.
• The fidelity of estimated responses depends on the number of uncorrelated noise sources.