DIGITAL TECHNOLOGIES FOR PREDICTING THE OPERATIONAL RELIABILITY OF GLASS-FIBER MECHANICAL STRUCTURES

This study provides a theoretical framework for the application of digital technologies to predicting the in-service reliability of glass-fibre-reinforced polymer (GFRP) engineering structures. The principal degradation mechanisms of GFRP composites are examined, including polymer matrix cracking,fibre matrix interfacial debonding, interlaminar delamination,local fibre failure, and progressive stiffness degradation.The effects of cyclic loading, temperature, moisture, vibration, and manufacturing defects on the evolution of structural condition are identified. A digital prognostic system is proposed as a closed-loop information and computational framework integrating a digital product passport, sensor-based monitoring, non-destructive testing, finite element modelling, a digital twin, and a prognostic module. The need for regular updating of the digital twin using in-service measurements and for accounting for the asset - specific loading history is substantiated. The integrated application of fibre-optic, piezoelectric, acoustic-emission, and strain sensors in conjunction with ultrasonic and thermographic inspection is shown to be appropriate. The potential of machine-learning techniques for sensor-data processing, diagnostic-feature extraction, and acceleration of prognostic calculations is examined. It is emphasised that such techniques should be applied in combination with physics-based models of composite deformation and failure.