Prognosis of Structural Damage Growth Via Integration of Physical Model Prediction and Bayesian Estimation

Damage diagnosis and prognosis play an important role in ensuring the safety of mechanical, aerospace, and civil structures. Most existing structural damage estimation methods are limited in only providing an estimate of the damage magnitude at the current time instance. Revealing the evolving path of structural damage is highly desirable in practice for prognosis and remaining useful life prediction. In this paper, we propose a dynamic data-driven hierarchical Bayesian degradation model, which takes advantage of both the physical finite element model and the data-driven Bayesian framework, to tackle the structural damage growth prediction. The damage growth trend can be efficiently and accurately estimated by Gibbs sampling. Systematic case analyses are performed to validate and demonstrate the effectiveness of the proposed method.