Research on Features and Localization of AE Signals from the Mechanical Body of Industrial Robots Based on Wavelet Denoising and EMD

Abstract

Abstract To reveal the propagation characteristics of acoustic emission(AE) signals within the mechanical body of an industrial robot, this study extracted the frequency features and wave velocities of AE signals when propagating on the interior and exterior surfaces.Subsequently,an AE source identification algorithm based on characteristic frequencies and time difference of arrival(TDOA) was proposed.AE sources were generated using Pencil-Lead Break(PLB),and signals were collected with piezoelectric sensors.Next,a denoising process was applied to the raw signals using a 4-layer wavelet decomposition method,and the empirical mode decomposition(EMD) method decomposed the denoised AE signals into independent modal functions(IMFs).The AE waveforms were then reconstructed using information from correlation coefficients and variance accounted for(VAF),enabling the extraction of frequency domain characteristics when AE signals propagated on the interior and exterior surfaces of the mechanical body.Frequency domain characteristics were extracted, and propagation speeds were calculated using the TDOA principle.A localization algorithm based on geometric relationships was developed.Results show exterior coating significantly affects AE propagation characteristics.Velocities on interior and exterior surfaces are 4463m/s and 4743m/s.The characteristic frequency ranges measured by the two sensors are 130kHz ~ 145kHz and 150kHz ~ 170kHz on the exterior surface, and 110kHz ~ 120kHz and 130kHz ~ 140kHz on the interior surface.The localization accuracy of the proposed AE source localization algorithm ranges from 0.2%~2.9%.