<title>Multi-damage detection with embedded ultrasonic structural radar algorithm using piezoelectric wafer active sensors through advanced signal processing</title>

Abstract

The embedded ultrasonic structural radar (EUSR) algorithm was developed by using piezoelectric wafer active sensor (PWAS) array to detect defects within a large area of a thin-plate specimen. EUSR has been verified to be effective for detecting a single crack either at a broadside or at an offside position. In this research, advanced signal processing techniques were included to enhance inspection image quality and detect multiple damage. The signal processing methods include discrete wavelet transform for signal denoising, short-time Fourier transform and continuous wavelet transform for time-frequency analysis, continuous wavelet transform for frequency filtering, and Hilbert transform for envelope extraction. All these signal processing modules were implemented by developing a graphical user-friendly interface program in LabVIEW. The paper starts with an introduction of embedded ultrasonic structural radar algorithm, followed with the theoretical aspect of the phased array signal processing method. Then, the mathematical algorithms for advanced signal processing are introduced. In the end, laboratory experimental results are presented to show how efficiently the improved EUSR works. The results are analyzed and EUSR is concluded to have been improved by using the advanced signal processing techniques. The improvements include: 1) EUSR is able to provide better image of the specimen under monitoring; 2) it is able to detect multi-damage such as several cracks; 3) it is able to identify different damage types.