Utilizing inverse uncertainty quantification techniques, structural health monitoring data can be integrated with damage progression models to form probabilistic predictions of a structure’s remaining useful life. However, damage evolution in realistic structures is physically complex. Accurately representing this behavior requires high-fidelity models which are typically computationally prohibitive. In this paper, high-fidelity fatigue crack growth simulation times are significantly reduced using a surrogate model trained via finite element analysis. The new approach is applied to experimental damage diagnosis data to form a probabilistic prediction of remaining useful life for a test specimen under mixed-mode conditions.
doi: 10.12783/SHM2015/299