Abstract
The MPC Omega model has become popular in recent years for the prediction of creep deformation. Successful predictions of the tertiary creep for a wide range of materials are available. The Omega model relates the strain as a linear function of the natural logarithm of strain-rate. It is assumed that the primary creep is a short-lived phenomenon and can be neglected. The Omega model is unable to predict the primary creep deformation. Often primary creep is a long-lived phenomenon and cannot be neglected. A mathematical modification can be performed to incorporate the primary creep curve in the Omega model. A common approach is by adding a work hardening function to the original constitutive model. Approaches using power, or exponential, or logarithmic work-hardening function are available. However, it is difficult to discern which function is the best for accurate prediction. In this study, the Omega model is modified to predict the primary and tertiary creep deformation curve by adding a hyperbolic tangent work hardening function. A metamodel incorporating the four modified Omega sub-models (power, exponential, logarithmic and hyperbolic tangent) is developed. The metamodel enables the determination of the most suitable model for a given material and avoids the force fit of a preselected model. Short, medium, and long-term creep deformation data for alloy P91 (pipe) and G91 (plate) at two isotherms of 600°C and 650°C are used to calibrate the metamodel. The data include five stress levels ranging from 70 to 160 MPa including creep life from 233 to 1.1 × 105 hrs. A detail calibration process is provided. A numerical analysis is performed to compare the four submodels. It is observed that the selection of the most suitable function depends on the loading condition and material properties. Based on the analysis, a recommendation to select the suitable work-hardening function to predict the primary and tertiary creep deformation curve is presented.