Most damage growth models require accurate stress intensity factor as well as model parameters for predicting damage growth. Depending on geometries and loading conditions, these models become complicated with additional model parameters. This paper shows that a simple model, such as the Paris model, can be used for complex geometries by compensating the error in stress intensity factor with the equivalent model parameters that are different from the true ones. Actual damage growth is simulated using the extended finite element method to model the effects of crack location and geometry on the relationship between crack size and stress intensity factor. The detection process of crack using structural health monitoring systems is modeled by adding random noise and a deterministic bias. The equivalent model parameters are then identified using the least-square-filtered Bayesian method, from which the remaining useful life is estimated. Using three examples, it is shown that the RUL estimates are accurate even when an inaccurate stress intensity factor is used.

This content is only available via PDF.
You do not currently have access to this content.