Zone-based reliability analysis on fatigue life of GH720Li turbine disk concerning uncertainty quantification

摘要

Probabilistic-based design for turbine disk can quantify risk and thus improve the reliability of the component. This paper presents a zone-based reliability framework in combination with uncertainty quantification for low cycle fatigue (LCF) life prediction of a turbine disk. In this investigation, a zone based modeling methodology is proposed instead of the traditional hot spot method, which accounts for all of the potential regions influencing the turbine disk's failure probability. The LCF life reliability of the whole turbine disk is achieved by jointing failure probability at each zone. The reliability framework developed involves the uncertainties from physical inputs and model parameters, in which geometry and model parameters are quantified using statistical method and Bayesian inference respectively, while the discretization error induced by finite element analysis is corrected using Richardson extrapolation method. Finally, the reliability analysis on the LCF life of a GH720Li turbine disk is performed considering the uncertainty quantification. It is demonstrated that the proposed zone-based reliability method in this paper is more accurate compared with the traditional hot spot method. (C) 2017 Elsevier Masson SAS. All rights reserved.