Validation of Model Order Assumption and Noise Reduction Method for the Impact Resonance Testing of Asphalt Concrete

The impact resonance test is a free vibration-based nondestructive test method that has been increasingly used in evaluation and characterization of asphalt concrete for the past two decades. The rheological modeling of the impact resonance test is conceptualized by a linear viscous damping mechanism having single degree of freedom whose equation of the motion is assumed to be second order. In this study, the second order equation of motion assumption in the modeling of the impact resonance test response was evaluated for asphalt concrete testing. A set of asphalt concrete specimens was tested with the impact resonance test, and the obtained signals at a range of temperatures were evaluated by means of the Hankel matrix method. The results showed that the assumption is violated for asphalt concrete testing especially at high temperatures, mainly due to the presence of noise in the obtained response. However, the Hankel method was employed to filter out the noise. It was seen that the assumption could be employed for asphalt concrete at a range of temperatures including high temperatures, provided that the filtering is performed on the obtained signal. The results also showed that the employed filtering procedure produced improvements for the impact resonance test material dependent responses, resonant frequency and especially damping ratio calculations.