Robust Implementation of Physical Regime Sensitivity Applied to Contrast Different Instability Experiments to Measure Strength in Extreme Conditions

Traditional numerical sensitivity analysis quantifies sensitivity to constitutive or global model parameters. Physical Regime Sensitivity (PRS) extends sensitivity analysis to a model's independent variables to reveal which regimes have the largest influence on an experiment, where sensitivity is measured with respect to a baseline, unperturbed simulation. Here PRS is robustly implemented in FLAG, an arbitrary Lagrangian-Eulerian (ALE) multiphysics code, and is demonstrated using a Preston-Tonks-Wallace (PTW) strength model in two tantalum experiments purported to exercise solid strength in extreme conditions: shock-driven Richtmyer-Meshkov Instabilities (RMI) and laser-driven Rayleigh-Taylor (RT) instabilities. These experiments employ complex loading, and interpretation is challenging without employing PRS or an alternative. In the dozens of parallel hydrodynamic simulations required for each variable in the PRS study, strength is perturbed in regimes of strain, strain rate, pressure, and temperature. State variable regimes that produce sensitivity are investigated in the simulations to reveal the underlying reasons for the strength sensitivity. The experiments show contrasting sensitivities to regimes of all the variables, which is interpreted in terms of their complementary role in covering the over-arching experimental space used for strength model calibration.