A numerical investigation of creep-fatigue life prediction utilizing hysteresis energy as a damage parameter

This paper explores the hypothesis that there exists an intrinsic material property, hysteresis damage energy at failure, which could be used as a creep-fatigue life prediction parameter. The connection between hysteresis energy and fatigue damage was introduced in the 1920’s by Inglis, but the use of hysteresis energy as a measure of damage was first presented by Morrow and Halford. Hysteresis energy shows promise in bridging the gaps associated with life prediction when the combination of both creep and fatigue scenarios are present. Numerical simulations which replicate experimental test configurations with 9Cr–1Mo steel were performed from which the hysteresis energy failure density (HEFD) could be calculated for each experiment. Taking the average of the HEFD values calculated for all of the experimental data as the parameter for failure (EIntrinsic), creep-fatigue life predictions were made using a simplistic hysteresis energy based method as well as the time fraction/cycle fraction method endorsed by ASME Code and compared to experimental results. A good correlation with experimental results was obtained for life predictions using hysteresis energy density as a damage parameter. An investigation of the interaction between creep damage and fatigue damage based on the hysteresis energy method was also performed and compared with the damage interaction diagram utilized by the ASME and RCC-MR design codes. The hysteresis energy based method proved easy to implement and gave improved accuracy over the time fraction/cycle fraction method for low cycle creep-fatigue loading.