High-resolution damage detection based on local signal difference coefficient model

Abstract

Probability reconstruction algorithm is a new promising tomography approach for the detection and monitoring of critical areas in a structure. In this algorithm, the correlation calculation is performed by capturing interrogation signals in different conditions to generate tomographic feature (i.e. signal difference coefficient) and linking the value with the presence of the potential defect. However, the way of signal difference coefficient in depicting defect is rough to some extent. Essentially, signal difference coefficient merely suggests how close the defect is away from the sensing path. The major reason is that the global signal is directly adopted without considering local information, which is equivalent to ignoring the fact that the existence of the defect always changes local signal rather than the global one. Under this limitation, signal difference coefficient restricts the resolution of the image. A new signal-processing technique is established to exploit the interrogation signal to improve the imaging performance in this article, which extracts local Lamb wave signal, uses local signal difference coefficient as a new input feature, and then employs it in probability reconstruction algorithm for damage identification. The efficiency of the proposed method is validated by some experiments.