查看文章

In this study, aiming at existing challenges in guided wave-based structure health monitoring (GWSHM) and vibration-based structure health monitoring (VSHM) methods, a hybrid method combining the advantages of these two methods is proposed to detect and localize damage on aluminium plate. Specifically, damage sensitive features are firstly extracted from guided wave and vibration, and then these features are input to a neural pattern recognition and a Backpropagation Neural Network (BPNN) to detect and localize the damage. Experimental results demonstrate that both the individual vibration and guided-wave methods are capable of achieving 100% accuracy in damage detection. However, in terms of damage localization, the GWSHM method achieves the best performance with an average Mean normalized distance (MND) value of 0.91cm and a minimum MND value of 0.66cm between the predicted …