Corrosion Detection/Quantification on Thin-Wall Structures Using Multi-Mode Sensing Combined With Statistical and Time-Frequency Analysis

In this paper, we present a multiple mode sensing methodology to detect active corrosion in aluminum structure utilizing the broadband piezoelectric wafer active sensors. This method uses ultrasonic Lamb wave complemented with the electromechanical impedance measurement to detect, quantify, and localize the corrosion progression in plate-like structures. The ultimate objective of this research is to develop in-situ multimode sensing system for the monitoring and prediction of critical aerospace structures that can be used during in-service period, recording and monitoring the changes over time. The test experiments were conducted on an aluminum plate installed with a five sensor network using 7-mm piezoelectric wafer active sensors. The corrosion was emulated as material loss of an area of 50mm 38mm on the other surface of the plate. Detection of corrosion and its growth was first conducted using the Lamb wave method in pitch-catch mode. The corroded area resulted in a thickness loss on the Lamb wave propagation and caused the amplitude and phase changes in the structural responses. The experimental data was first evaluated by the statistics-based damage indicator using root mean square deviation. Though the damage indicator is able to detect the presence of the corrosion and identify the corrosion location quantitatively, it failed in giving the right indication of corrosion development. A more corrosion signal processing based method, the cross time-frequency analysis, was proposed and used to analyze the phase characteristics of the data set. This cross time-frequency analysis was found more reliable and precise for detecting the corrosion progression compared with the damage indicator method.