Prediction of impact-damage growth in GFRP plates using particle filtering algorithm

Abstract

With increasing use of fiber reinforced polymer (FRP) composites in several industrial applications, structural health monitoring and prognosis have become an extremely critical task in recent years. Accurate health prognosis ensures system reliability and aids in estimating the remaining-useful-life (RUL) which in turn reduces repair or replacement costs. In this paper, a framework for the estimation of impact damage propagation in GFRP plates is proposed which utilizes a physical model based on Paris’ law and data obtained from inspection of GFRP specimens by optical transmission scanning (OTS) technique. Advanced signal processing and feature extraction is performed to quantitatively characterize the delamination size from the OTS data. A Bayesian method based on particle filter update has been implemented to estimate the model parameters by taking observed data into account and accurate RUL prediction of the GFRP samples subjected to impact damage has been achieved. Results demonstrate feasibility and potential of the proposed approach as a robust reliability analysis technique of GFRP laminar plates.