abstract
- An analysis of damage evolution of aged glass fiber epoxy composites subjected to bending loading was performed by the combined use of acoustic emission (AE) and self-monitoring ability enabled by the incorporation of multiwall carbon nanotube (MWCNT) networks. The acoustic signal analysis was conducted using a classification methodology based on the k-means algorithm and principal component analysis (PCA). This approach enabled the detection of failure modes including matrix fiber-matrix separation, matrix failure, fiber rupture, and delamination. The aged samples retained their piezoresistive response, and by monitoring changes in electrical resistance along with key acoustic emission (AE) parameters, it was possible to track specific failure events throughout the damage evolution of the composites. In order to monitor the damage evolution of samples with piezoresistive capability, and similarly to the definition of the sentry function (SF), a pseudo-SF was proposed, but by utilizing the variation in electrical resistance instead of the cumulative acoustic energy as the SF is normally defined. Damage assessment of the failed specimens was performed through the use of x-ray computed tomography and scanning electron microscopy (SEM). The collective use of piezoresistivity and AE techniques provides a reference for the structural health monitoring (SHM) of fiber-reinforced materials in marine applications. © 2025 Society of Plastics Engineers.