Terahertz data analytics-based bonding interface damage characterization in a multilayer structural composites under cyclic loading

Abstract

Adhesively-bonded composite structures are under various environmental loads during their service life. Therefore, it is imperative to monitor the evolution of the bonding interfaces under loading to evaluate the bonding properties of the composite structures. A terahertz (THz) quantitative characterization method is proposed based on the scale invariant feature transform (SIFT) image matching algorithm. The terahertz data for the bonding interface before and after loading were obtained using a terahertz time-domain spectroscopy system along with the flight time imaging of the bonding interface. The imaging results of the bonding interface before and after loading were matched by SIFT image processing technology, and the THz flight time difference was obtained using differential calculations to characterize the thickness change of the bonding interface. Furthermore, the initiation and development of interface damage were studied, and the changes in the bonding interface thickness and debonding damage state under the cyclic loading tests were analyzed, providing information on the growth of the damaged area and damage severity. The results show that the proposed method can ultimately realize the characterization of the thickness changes and the evolution of the debonding damages at the interface of the bonded composite structures under cyclic stress, thus circumventing the shortcomings of the conventional damage-evolution characterization methods. The findings can also provide helpful insights into the effect of stress load on the bonding interface damage evolution for the studied system.