Lay-up optimisation of fibre metal laminates based on fatigue crack propagation and residual strength

•A design optimisation procedure for fibre metal laminates is presented.•Prediction methods are used to evaluate the fatigue crack propagation behaviour and the residual strength of a laminate.•A fitness approximation is performed to speed up the optimisation procedure.•Genetic algorithm is applied...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Composite structures 2015-06, Vol.124, p.77-87
Hauptverfasser: Şen, I., Alderliesten, R.C., Benedictus, R.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•A design optimisation procedure for fibre metal laminates is presented.•Prediction methods are used to evaluate the fatigue crack propagation behaviour and the residual strength of a laminate.•A fitness approximation is performed to speed up the optimisation procedure.•Genetic algorithm is applied as optimisation algorithm. A methodology to design and optimise fibre metal laminates for improved fatigue and damage tolerance properties is presented. The lay-ups are defined in a systematic manner where the number and thickness of metal layers are varied and the lay-ups are divided into grades in which the amount and orientation of the fibre plies in the fibre layers are defined. The optimisation procedure is implemented with genetic algorithms and the lay-ups are designed such that the fatigue crack propagation or residual strength criteria is satisfied. The design criteria are evaluated using prediction methods and fitness approximations of these prediction methods. The latter evaluation aims to speed up the optimisation procedure. The functions of the fitness approximation are verified against the prediction methods and the design solutions of both evaluation methods are in compliance with each other. In conclusion, the procedure managed to find the optimal solutions within the design space while an improvement in computation time is achieved with the use of fitness approximation.
ISSN:0263-8223
1879-1085
DOI:10.1016/j.compstruct.2014.12.060