Analysis of impact sound waveform by answer-in-weights neural network

A method to estimate the elastic moduli of composite material from the impact sound was proposed. For estimating the degree of fatigue of the material, it is important to detect long-term components as well as periodic components. We propose to use answer-in-weights networks for estimating the periodic and long-term components of the damping waveform, simultaneously. To show the effectiveness of the method, we performed simulations on the damping vibration data occurring due to the impact sound of the composite material.