A Coupled Computational Welding Mechanics and Physics-Based Damage Modeling for Prediction of Remaining Useful Life in Welded P91 Alloy

Loss of creep resistance in post-weld P91 alloy occurs mainly due to the change in grain size as well as residual stress from the welding process. Post-weld heat treatments can partially improve the creep life, however, it remains important to determine the remaining useful life (RUL) particularly in the heat-affected zone under actual service operating conditions. Most creep damage models have focused on the short-term creep response at a relatively high temperature and stress, where the deformation mechanism is governed by power-law creep (PLC). However, under the actual service temperature and stress (i.e. about half of the melting point) grain-boundary sliding (GBS) is the dominant deformation mechanism that contributes to the creep life. In this paper, a validated deformation mechanisms map (DMM) using low temperature creep strain accommodation processes i.e. GBS, is developed for P91 alloy that predicts the creep rates over a wide range of temperature and stress including those arising under in the actual service conditions. These creep rates are further utilized into a microstructure-based creep damage model for accurate life prediction. A 3D transient computational welding mechanics (CWM) modeling of a pipe in a super-critical water loop, predicts the thermal, microstructure and stress state from welding. It also determines the coarse and fine grain heat affected zone (CGHAZ & FGHAZ). The CWM results are coupled with physics-based creep damage modeling to predict the RUL under the actual service conditions considering the welding residual stress and microstructure states