Abstract:
This work is aimed to address the random nature of the progressive damage process (PDP) in laminated composite structures for enhancement of engineering solutions and decision-making based on damage tolerance assessment (DTA). A general approach integrating input characterization, output validation, and corresponding modeling solutions is suggested for understanding and quantification of inevitable randomness of multi-step PDP in laminated composites and components. A general modeling approach of probabilistic PDP is developed based on system definition of laminated composites consisting of elements with random properties and random load distribution between them. The approach is implemented in form of a parametric model based on Finite Element Analysis (FEA) coupled with statistical simulations. Damage initiation and damage growth are captured through cohesive-element FEA. FEA/probabilistic coupling is performed through developed Python/MATLAB software. Robust demonstration of developed modeling capabilities is shown on examples of thick laminated composite beam with a network of multiple potential interlaminar damages. Quick convergence with respect to number of statistical simulations is shown. Preliminary qualitative comparison with similar available experimental results indicates good match between observed and predicted damage processes. Challenges of probabilistic characterization and validation are listed and discussed. Finally, directions of follow-up expansions of the developed capabilities are proposed for future efforts.