Observation-infused simulations of high-speed boundary-layer transition

High-speed boundary-layer transition is extremely sensitive to the free-stream disturbances which are often uncertain. This uncertainty compromises predictions of models and simulations. To enhance fidelity simulations, we directly infuse them with available observations. Our methodology general can be adopted any simulation tool, herein demonstrated using direct numerical An ensemble variational (EnVar) optimization performed, whereby determine upstream flow that optimally reproduces The cost functional accounts for our relative confidence in model observations, judicious choice members improves convergence reduces prediction uncertainty. We demonstrate observation-infused at Mach 4.5. Without prior knowledge condition, only observations wall pressure isolated locations from an independent computation (true flow), all relevant inflow identified. then evaluate entire field, beyond original limited interpret simulations consistently data vice versa. predicted compares favorably true `unknown' state, discrepancies analyzed detail. also examine impact weighting Improved inverse problem accuracy amplitudes phases demonstrated, explained by aid a simple example two-dimensional unstable, chaotic convection.