In Situ Damage Decoupling and Prognostic Maintenance Decision-making for Aerospace Composite Materials
DescriptionDamage mechanisms in Carbon Fiber Reinforced Plastics (CFRP) composite materials are complex which include matrix micro-cracking, ply delamination, and fiber breakage. Current damage identification methods in CFRP composites usually focus on a single damage mode which is not practical in real world scenarios. This proposal aims to decouple the two vital damage modes which are matrix micro-cracking and ply delamination by analyzing the ultrasonic guided wave signals collected from the CFRP composite specimens. After decoupling, the delamination will be identified and localized using a deep learning-based classification approach. The matrix micro-cracking will then be quantified using a deep learning-based regression approach. By quantitatively evaluating the damage intensity of both ply delamination and matrix micro-cracking, a probability distribution of the remaining service life for the composite structures can be generated. Based on this distribution, a decision making under uncertainty model will be established for prognostic maintenance management of these composite structures.
|Effective start/end date
|1/04/23 → …