A new non-linear impact resonance acoustic spectroscopy algorithm: FANSIRAS

Abstract

Nonlinear hysteretic techniques have demonstrated their suitability for damage detection in mesoscopic materials. These methods are mainly based in the evaluation of the frequency and damping shifts with the increasing strain amplitude under dynamic conditions. A deep understanding of the impact reverberation phenomena has been achieved in order to introduce a novel signal processing approach called FANSIRAS (Flipped Accumulative Non-linear Single Impact Resonance Acoustic Spectroscopy). This brand new procedure respects the underlying phenomena, composed of both an exponentially decaying signal and a time varying phase signal, which results in the feasibility of using a single impact to evaluate the non-linear features of damaged materials. The traditional acoustic spectroscopy technique, NIRAS (Non-linear Impact Resonance Acoustic Spectroscopy), and FANSIRAS have been analyzed and compared when providing quantitative information related to the degree of damage. In this work, both techniques were applied to thermal damaged concrete specimens showing equivalent results in the hysteretic parameters calculation and clear differences between pristine and damaged states. FANSIRAS algorithm takes advantage of the appropriate processing of a single reverberation signal to estimate equivalent acquisitions to those obtained from NIRAS with several blow impacts. FANSIRAS simplicity and robustness may be important in several engineering applications © 2018 Acoustical Society of America.