In the present study, a general probabilistic design framework is developedfor cyclic fatigue life prediction of metallic hardware using methods thataddress uncertainty in experimental data and computational model. Themethodology involves (i) fatigue test data conducted on coupons of Ti6Al4Vmaterial (ii) continuum damage mechanics based material constitutive models tosimulate cyclic fatigue behavior of material (iii) variance-based globalsensitivity analysis (iv) Bayesian framework for model calibration anduncertainty quantification and (v) computational life prediction andprobabilistic design decision making under uncertainty. The outcomes ofcomputational analyses using the experimental data prove the feasibility of theprobabilistic design methods for model calibration in presence of incompleteand noisy data. Moreover, using probabilistic design methods result inassessment of reliability of fatigue life predicted by computational models.
展开▼
