Development of a Variable-Frequency Hammering Method Using Acoustic Features for Damage-Type Identification

Hammer testing, a nondestructive testing method, has been demonstrated to provide information on structural damage. One of the biggest challenges with this method is simultaneous identification surface, internal, and composite damage (consisting both surface internal damage) in complex environment, such as post-disaster. A based variable-frequency hammering proposed solve problem. The importance feasibility using impact hammers generated acoustic data identify multiple types concrete structures are presented. First, type feature was extraction selection obtained from hammering. Second, recognition model established support vector machine four occurring simultaneously same member specimens, including two within 20 mm. Finally, verified experimentally. This exhibited good performance, an accuracy 97.8%; moreover, ensures that dimensionality remains unchanged while increasing effective data.