Exploratory Study of Impulse Response for Locating Defects Using a Microphone and Empirical Mode Decomposition
Sikandar Sajid & Luc Chouinard
What the paper says
This research presents a novel procedure of using stationary waves generated with the impulse-response test using semi-noncontact measurements and relatively high-frequency bandwidths. The time histories of vibrations are recorded with a condenser microphone over a regular square grid at the surface of a concrete plate. The analyses were performed on the normalized time-domain response at each test location. Using the impact-normalized time-domain responses, oscillatory modes are obtained through empirical mode decomposition. The first oscillatory mode of each normalized response is converted back to the frequency domain, and its maximum amplitude is shown to be an accurate defect indicator. The defects are effectively delineated spatially, and their intensity is estimated by using extreme value outlier analysis. The physical basis for the proposed procedure is demonstrated by using a finite-element model that indicates that stationary acoustic waves are generated by an impact above the concrete surface. An experimental specimen with simulated defects in the form of shallow and deep delamination, debonding, and honeycomb was used to validate the proposed procedure. Using the proposed experimental and the new data analysis procedure, the defect indicator was able to detect the delamination at two different depths, debonding, and honeycomb in the lab-scale concrete plate. The level of accuracy of the damage detection was comparable to that of more refined yet time-consuming ultrasonic shear-wave echo method. The defect indicator calculated with the proposed signal processing algorithm was found to be robust to the ambient noise that can influence the measurements with the condenser microphone.
Evidence weight
Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40
| F · citation impact | 0.50 × 0.4 = 0.20 |
| M · momentum | 0.50 × 0.15 = 0.07 |
| V · venue signal | 0.50 × 0.05 = 0.03 |
| R · text relevance † | 0.50 × 0.4 = 0.20 |
† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.