Ensemble-Kalman-Filter-Based Data Assimilation for Shock Wave/Boundary-Layer Interaction

An ensemble-Kalman-filter-based data assimilation (DA) framework is developed to optimize turbulence models for shock-wave/boundary-layer interactions (SWBLIs), using the [Formula: see text] shear stress transport (SST) model as a representative case. Comparing to existing methods which rely on case-specific tuning or high-fidelity data, the ensemble Kalman filter DA systematically tunes SST model constants using limited experimental data to enhance separation predictions in SWBLIs. The results show that DA-driven modifications improve key turbulence quantities in the separation zone, leading to more accurate wall pressure distributions and refined predictions of separation and reattachment locations. Furthermore, the DA-optimized SST model not only exhibits strong generalization capability across various Reynolds numbers and interaction strengths, but it also performs reasonably well when being extended from compression ramp to impinging shock interactions. These findings demonstrate the broader potential of DA-based turbulence model optimization in supersonic SWBLI modeling.