logo SBA

ETD

Archivio digitale delle tesi discusse presso l’Università di Pisa

Tesi etd-06282016-105916


Tipo di tesi
Tesi di laurea magistrale
Autore
DI MARCO, VINCENZO
Indirizzo email
vincenzodimarco@outlook.com
URN
etd-06282016-105916
Titolo
NUMERICAL STRATEGIES FOR THE PREDICTION OF FATIGUE CRACK PROPAGATION IN COLD-EXPANDED HOLES
Dipartimento
INGEGNERIA CIVILE E INDUSTRIALE
Corso di studi
INGEGNERIA AEROSPAZIALE
Relatori
relatore Ing. Boni, Luisa
relatore Ing. Fanteria, Daniele
Parole chiave
  • stress intensity factor
  • residual stress fields
  • fatigue crack propagation
  • cold-expansion
  • cold working
  • XFEM
Data inizio appello
19/07/2016
Consultabilità
Completa
Riassunto
Split sleeve Cold eXpansion (CX) is a common technique to improve the fatigue behaviour of open holes as well as fastener holes in fatigue critical joints of aeronautical metallic structures. Many test evidences in the past have demonstrated the capability of cold expansion to provide fatigue life enhancement, such as longer inspection intervals or a general increased operational lifetime of the airframe. Thus for several decades it has been applied in high loaded aeronautical components to extend their fatigue lives, taking into account the beneficial effect of the compressive residual stress induced by the CX technology on fatigue initiation. However the healthy effects of the CX process are still not commonly considered in the design phase of an aircraft. In fact, in order to predict the crack growth in the presence of residual stresses, it is essential to calculate accurately the Stress Intensity Factor (SIF) related to a crack growing in 3D residual stress fields. The development of a predicting capability in this case remains still challenging, especially if the crack is located in regions with high stress gradients. For this purpose, a dedicated numerical strategy has been elaborated at UNIPI to solve the issues connected to the SIF evaluation and to model crack propagation in complex residual stress fields. The discontinuity has been modelled using both innovative methods (e.g. XFEM) and traditional approaches. A massive use of in-house Python codes has simplified and speed up the phases of model generation and data extraction. Finally the obtained results have been compared and validated through a focused experimental activity.
File