Artificial Intelligence Assisted Residual Strength and Life Prediction of Fiber Reinforced Polymer Composites

Abstract

With the increased use of composite materials, researchers have developed many approaches for structural and prognostic health monitoring. Broadband Dielectric Spectroscopy (BbDS)/Impedance Spectroscopy (IS) is a state-of-the-art technology that can be used to identify and monitor the minute changes in damage initiation, accumulation, interactions, and the degree of damage in a composite under static and dynamic loading. This work presents a novel artificial neural network (ANN) framework for fiber-reinforced polymer (FRP) composites under fatigue loading, which incorporates dielectric state variables to predict the life (durability) and residual strength (damage tolerance) from real-time acquired dielectric permittivity of the material. The findings of this study indicate that this robust ANN-based prognostic framework can be implemented in FRP composite structures, thereby assisting in preventing unforeseeable failure.