A model to predict the evolution of pitting corrosion and the pit-to-crack transition incorporating statistically distributed input parameters

摘要

A model based on deterministic equations with statistically distributed input parameters has been developed for simulating the evolution of the pit depth distribution at different exposure times and the percentage of pits that transform to stress corrosion cracks. The model has been applied to the specific case of steam turbine disc steel exposed to a range of environments under applied stress. With preliminary fitting at one exposure time, the simulation not only reflects the trends in the experimental measurement but also the model, uniquely, reproduces the statistical variability or "noise" associated with the measurements.