ABSTRACT
In addition to the techniques mentioned above, the Bayesian approach is also popular, and has been widely applied to SHM data by many researchers. The major advantage is that it enables the inclusion of existing design knowledge as prior information. Therefore, even without extensive life-cycle data, the updated model can still provide a reasonable prediction and reflect the bias of design knowledge as well as including the remaining uncertainty in the prediction. Past work in study of applying Bayesian technique into structural fatigue life updating can be generally divided into two groups: one focuses on updating structural damage model from crack inspection record with fatigue loads being deterministically decided (Straub 2009). The other group tries to quantify fatigue load uncertainty from strain measurement with a design stage life prediction model (Nielsen et al. 2011).
