Probabilistic life prediction of structures under fatigue and creep

A simulation-based methodology for creep-fatigue life prediction and risk assessment is developed in this study for structural components. The proposed method considers both creep and fatigue failure mechanisms, the interaction effect between fatigue and creep, and the interaction effect between different loading levels. The predicted life includes both crack initiation and crack propagation.; The proposed method considers the basic uncertainties in the procedure of creep-fatigue life prediction. Random loading histories are simulated with the Monte Carlo method to consider the uncertainties in the loading. Material properties are also assumed to be random variables. The scatter in crack growth rate is modeled. The predicted probabilistic creep-fatigue life considering all these uncertainties is observed to follow a lognormal distribution.; Continuum damage mechanics is used to develop the model of damage evolution in composite materials. The proposed damage model is capable of capturing the unique characteristics of damage accumulation in composites. This model is used for damage evaluation and life prediction in composites subjected to time-variant loading. The proposed model is verified with numerical examples and data available in the literature.; Creep behavior of composite materials is studied. A new method is developed to consider the effect of stochastic fiber fracture on the creep failure probability. The importance of different random variables on the failure probability is investigated through sensitivity analysis. The obtained sensitivity information provides guidelines for data collection.; Risk assessment of the structure can be performed with the probabilistic life prediction method developed above. Appropriate scheduling of inspection and maintenance can be made based on the risk assessment to ensure sound performance of the structure during its service life. The inspection results can be used to update the failure probability using the Bayesian approach. Reliability-based optimization is combined with the Bayesian approach to develop an optimal inspection strategy with minimal total cost while fulfilling the requirement of reliability.