Effect of Number of Paris Law Parameters on Bayesian Inference for Remaining Useful Life Estimation

摘要: Information on damage sizes obtained from structural health monitoring (SHM) can be used to estimate remaining useful life (RUL). Damage growth information may also reduce uncertainty in the material properties that govern propagation for structure being monitored, turning aircraft into flying fatigue laboratories. These are often widely distributed between nominally identical structures because of manufacturing variability and aging effects. The reduced characteristics reduces turn prediction RUL monitored component. It help anticipating similar components. Bayesian inference has been progressively reducing structure-specific parameters spite noise bias sensor measurements. However, approach updating we using here computationally intensive, particular when it comes multiple variables. In this paper compare two characterize crack Paris law only one taking advantage correlation two. We find there is little gained by both parameters.