Experimental characterisation and computational modelling of cyclic viscoplastic behaviour of turbine steel - Université de Lorraine Accéder directement au contenu
Article Dans Une Revue International Journal of Fatigue Année : 2019

Experimental characterisation and computational modelling of cyclic viscoplastic behaviour of turbine steel

Résumé

Fully reversed strain controlled low cycle fatigue and creep-fatigue interaction tests have been performed at ±0.7% strain amplitude and at three different temperatures (400°C, 500°C and 600°C) to investigate the cyclic behaviour of a FV566 martensitic turbine steel. From a material point of view, the hysteresis mechanical responses have demonstrated cyclic hardening at the running-in stage and subsequent, hysteresis cyclic softening during the rest of the material life. The relaxation and energy behaviours have shown a rapid decrease at the very beginning of loading followed by quasi-stabilisation throughout the test. A unified, temperature- and rate-dependent visco-plastic model was then developed and implemented into the Abaqus finite element (FE) code through a user defined subroutine (UMAT). The material parameters in the model were determined via an optimisation procedure based on a genetic solver. The multi-axial form of the constitutive model developed was demonstrated by analysing the thermomechanical responses of an industrial gas turbine rotor subjected to in-service conditions. Asub-modelling technique was used to optimise the FEA. A 2D global model of the rotor with a 3D sub-model of the second stage of the low pressure turbine were then analysed in turn. The complex transient stress and accumulated plastic strain fields were investigated under realistic thermo-mechanical fatigue loading (start-up and shut-down power plant loads). The sub-modelwas then used for local analysis leading to identification of potential crack initiation sites for the presented types of blade roots.
Fichier principal
Vignette du fichier
LEM3_IJF_2019_RAE.pdf (5.45 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02317429 , version 1 (16-10-2019)

Identifiants

Citer

Yaroslav Rae, Adil Benaarbia, Jeremy Hugues, W. Sun. Experimental characterisation and computational modelling of cyclic viscoplastic behaviour of turbine steel. International Journal of Fatigue, 2019, 124, pp.581-594. ⟨10.1016/j.ijfatigue.2019.01.022⟩. ⟨hal-02317429⟩
26 Consultations
44 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More