Community Research and Development Information Service - CORDIS


This work presents a non-linear method for the modelling and identification of fatigue crack growth (FCG). The method is based on the powerful and well-known non-linear least squares technique for the parameter estimation. The relevant algorithms and its computational efficiency are also fully developed and proved. A case study with the Virkler data is performed. Several FCG laws are examined and the present non-linear methodologies are implemented and compared, where possible. The new parameter estimates, in combination with Monte Carlo numerical simulations, provide realistic predictions of the crack length distributions and generally lead to improved structural reliability assessments.

Additional information

Authors: MOUSSAS V C, University of Patras, Computer Engineering Department (GR);SOLOMOS G P, JRC Ispra (IT);LUCIA A C, JRC Ispra (IT)
Bibliographic Reference: Paper presented: 2nd International Conference on Computational Stochastic Mechanics, Athens (GR), June 13-15, 1994
Availability: Available from (1) as Paper EN 38554 ORA
Record Number: 199411285 / Last updated on: 1994-12-06
Original language: en
Available languages: en