Extraction of healthy part using two acoustic characteristics for defect detection by non-contact acoustic inspection method

Abstract

We have studied a non-contact nondestructive acoustic inspection method to detect internal defects of concrete structures using a long range acoustic device (LRAD) and a scanning laser Doppler vibrometer (SLDV). As an actual concrete structure, we measured the concrete floor slab of a huge bridge from a distance of about 33 m. And an internal defect of concrete was detected and visualized. Next, for a railroad tunnel, measurements were made from a distance 5-7 m. In data process, a healthy part of concrete was statistically analyzed. The difference in concrete state between a ceiling and a sidewall was evaluated by two acoustic feature quantities calculated from our measured data. We also measured the strength of concrete by Schmidt hammer and the elastic wave velocity. Both acoustic and physical feature quantities show a tendency of the same phenomenon. Obtaining the state of concrete, such as hardness, is to evaluate a healthy part of concrete. It can be done by using two acoustic feature quantities such as vibrational energy ratio and spectral entropy obtained from our measurements. In addition, in order to detect internal defects of concrete, it is important to grasp the distribution state of acoustic features quantities for a concrete healthy part.