Low-Velocity Impact Localization on a Honeycomb Sandwich Panel Using a Balanced Projective Dictionary Pair Learning...

Area of Science:

• Materials Science and Engineering
• Structural Health Monitoring
• Sensor Technology

Background:

• Carbon-fiber aluminum honeycomb sandwich panels are susceptible to low-velocity impacts, compromising structural integrity and performance.
• Rapid stress wave damping in these panels complicates impact localization using sensor networks.
• Accurate impact detection is vital for ensuring the reliability of structures utilizing these materials.

Purpose of the Study:

• To develop and validate a precise method for locating low-velocity impacts on carbon-fiber aluminum honeycomb sandwich panels.
• To address the challenges posed by signal attenuation and variation in sensor data for impact localization.
• To enhance structural health monitoring capabilities for sandwich panel structures.

Main Methods:

• Utilized fiber Bragg grating sensors for impact detection.
• Implemented a projective dictionary pair learning algorithm with structural sparse representation.
• Divided the panel into sub-areas, training separate dictionaries and grouping sensors as main or auxiliary.
• Introduced a balancing weight factor to optimize sensor contribution and mitigate poor signal quality effects.

Main Results:

• Achieved a 96.7% impact positioning accuracy on a 300 mm × 300 mm × 15 mm sandwich panel.
• Demonstrated an average positioning error of 0.85 mm.
• Validated the effectiveness of the balancing weight factor in improving localization accuracy.

Conclusions:

• The proposed method offers a highly accurate and reliable solution for impact localization in carbon-fiber aluminum honeycomb sandwich panels.
• The technique is suitable for structural health monitoring applications, providing critical data for damage assessment.
• The developed algorithm effectively overcomes signal quality issues, enhancing localization precision.