Lamb Wave Propagation-based Damage Identification for Quasi-isotropic CF/EP Composite Laminates Using Artificial Neural Algorithm: Part II--Implementation and Validation

摘要

Active transducer networks using distributed piezoelectric actuator/sensor were designed in terms of a concept of'Standard Sensor Unit' (SSU). Functionally integrating the artificial neural networks well-trained by Damage Parameters Database (DPD) developed in Part I, an active online structural health monitoring (SHM) system was configured on a VXI platform, which was then validated by quantitatively identifying hole-type defects in quasi-isotropic [0/45/-45/90]_s CF/EP (T650/F584) composite laminates. The system has exhibited excellent ability to quantitatively assess the damaged parameters, including presence, location, geometric identity, and orientation. Additionally, the reliability and performance of the SHM system on the inherent network configurations, such as architecture, training pattern, training function, and distribution of transducers, were also evaluated.