Introduction
Structural Health Monitoring (SHM) is essential for maintaining the integrity of critical infrastructure and industrial components. Ultrasonic Lamb waves are widely used for damage detection in plate-like structures due to their ability to propagate over long distances and interact with defects. However, conventional imaging methods, such as Delay-and-Sum (DAS), often suffer from "ghost" artifacts—visual noise that can obscure actual defects, particularly in multi-flaw scenarios. This paper presents a novel hybrid imaging algorithm that combines geometric four-point arc localization with probabilistic weighting to enhance detection accuracy and suppress artifacts.
Methodology
The proposed approach integrates three core components to improve imaging quality:
- Time Reversal (TR) Theory: This principle utilizes acoustic reciprocity. By recording a scattered signal and re-emitting it in reverse, the wave energy focuses back on the defect, which acts as a secondary source. The deviation between the original and reconstructed signals is quantified as a Damage Index (DI).
- Four-Point Arc Location: Unlike standard triangulation, this method employs four transducers arranged in a square. By solving the time-of-flight equations across these points, the algorithm precisely determines the propagation time from the transducer to the defect.
- Probability Weighting: To eliminate artifacts, a spatial filter is applied. A distribution function assigns higher weights to areas near the sensing paths where damage is statistically most likely to occur, effectively masking noise and highlighting the true defect location.
Experimental Setup
The effectiveness of the algorithm was validated using 6061 aluminum plates. The experimental system included a function generator, a power amplifier, an oscilloscope, and a PC for signal processing. An 8-probe transducer network was used to create multiple sensing paths across the plate. Damage was simulated using 15 mm diameter circular through-holes in both single-flaw and double-flaw configurations.
Key Results
The proposed method demonstrated significant improvements over the industry-standard Time-Reverse DAS method:
- Localization Accuracy: For a single flaw, the location error was reduced from 19.04 mm (DAS) to just 5.00 mm. In double-flaw tests, errors were kept between 15–17 mm, compared to 29–34 mm for DAS.
- Imaging Quality: Qualitative heatmaps showed a near-total elimination of the central artifact "cloud" that typically plagues DAS images, providing a much clearer visual representation of the defects.
Conclusion
By combining the geometric precision of the four-point arc method with the noise-suppression capabilities of probability weighting, this research offers a robust solution for high-fidelity damage imaging in plate structures. Future work will focus on optimizing wave velocity measurements and extending the method to complex composite materials.
Summary updated