Invariant Procedure for Error Sensitivity Analysis Applied to Cyclic Creep Modelling

Cyclic creep, also known as ratcheting, is a progressive accumulation of inelastic strain under cyclic stress-controlled loading. This mechanical effect is of great importance for numerous applications. In engineering practice, phenomenological models of cyclic creep are calibrated against a limited set of macroscopic test data. Since the testing results are prone to systematic and non-systematic experimental errors, the impact of experimental errors on the quality of simulation has to be analysed. A simple inspection procedure is demonstrated and tested. Based on the Monte Carlo computations, it allows for analysis of error propagation through the simulation cycle. The focus of the paper is on the independence of the procedure from the chosen model parametrization. For demonstration purposes, cyclic creep of VT6 alloy is simulated. The corresponding macroscopic constitutive equations are based on the second Ohno-Wang model, combined with refined rule of isotropic hardening. Two different parametrizations are introduced to show that the procedure predicts the same results for both of them.