Initial Arrival Time Identification Utilizing Wavelet Transform and Akaike Information Criterion for Locating Acoustic Emission Source in a Thin Plate

Acoustic emission (AE) method enables real-time monitoring of damage initiation and progression. Recently, AE analysis using machine learning has become widely popular; however, the source location is often located manually to ensure reliability accuracy. Therefore, it desirable that detection fully automated with a high This study proposes novel arrival time identification for can accurately automatically locate sources. First, wavelet transform applied an signal extract coefficient specific frequency. Subsequently, Akaike information criterion transient identify initial wave time. The localized accuracy compared conventional methods: visual identification, S0 mode at coefficients, automatic peak AIC identification. proposed reduces number events incorrect detections. Moreover, correct rate increases by 1.5 times normal AIC. In addition, approximately 30 faster than excellent in terms both speed analysis. Lastly, we verify be effectively anisotropic materials.