A reliability-based framework for fatigue damage prognosis of composite aircraft structures

Abstract The extensive use of lightweight composite materials in composite aircraft structures drastically increases the sensitivity to both fatigue- and impact-induced damage of their critical structural components during their service life. Within this scenario, an integrated hardware–software system that is capable of monitoring the composite airframe, assessing its structural integrity, identifying a condition-based maintenance, and predicting the remaining service life of its critical components is therefore needed. As a contribution to this goal, this paper presents the theoretical basis of a novel and comprehensive probabilistic methodology for predicting the remaining service life of adhesively bonded joints within the structural components of composite aircraft, with emphasis on a composite wing structure. Non-destructive evaluation techniques and recursive Bayesian inference are used to (i) assess the current state of damage of the system and (ii) update the joint probability distribution function (PDF) of the damage extents at various locations. A probabilistic model for future aerodynamic loads and a damage evolution model for the adhesive are then used to stochastically propagate damage through the joints and predict the joint PDF of the damage extents at future times. This information is subsequently used to probabilistically assess the reduced (due to damage) global aeroelastic performance of the wing by computing the PDFs of its flutter velocity and the velocities associated with the limit cycle oscillations of interest. Combined local and global failure criteria are finally used to compute lower and upper bounds for the reliability index of the composite wing structure at future times.