Neural-Network-Based Ultrasonic Inspection of Offshore Coated Concrete Specimens

A thin layer of protective coating material is applied on the surface offshore concrete structures to prevent its degradation, thereby extending useful life structures. The main reasons for reduction in capability layers are loss adhesion and flattening wall. Usually, state monitored setting water immersion using ultrasonic inspection methods, method still needs improvement terms speed accuracy. In this study, pulse echo was used a test coated specimens, continuous wavelet transform (CWT) with complex Morlet wavelets implemented define received waveforms’ time flight instantaneous center frequency. These allow one evaluate thickness during immersion. Furthermore, phases reflected echoes at CWT local peaks were computed combination Hilbert transforms (HT) wave parameters derived from CWT. addition, three relative also train deep neural networks (DNN), including frequency ratio, magnitude phase difference. With use waveform DNN, debonded detection accuracy our 100%.