A Mode Based Reduced Order Model for Supersonic Store Separation

Abstract

Successful supersonic store separation has historically been, and continues to be, a challenging task. Processes currently used for identifying potential failure conditions of store separation are both resource intensive and time consuming. It is this time consuming nature of store separation modeling which currently limits the possibility for improving separation trajectory through use of store design optimization and flight controller training. To address the challenges associated with store separation modeling, this work presents a modeling method which is capable of making high fidelity predictions for both surface pressure and shear stress distributions at minimal computational cost. To build this model, three supersonic Mach numbers were selected at 1.2, 1.4, and 1.6 for fixed wing store separation simulations. Calculated surface pressures and shear stress distributions were recorded at equal time intervals. Load distributions were analysed with Proper Orthogonal Decomposition (POD) and reduced to a subspace of 64 modes. Kriging interpolation was then used along side the equations of motion to predict store trajectories at Mach numbers of 1.3 and 1.5. Comparison between POD reduced order model (ROM) and CFD simulation show that the resulting model was able to not only make accurate trajectory predictions but also make close predictions for surface pressure and shear stress distributions. By using the ROM, computational run times were reduced from 6 hours to 1 minute while retaining a sufficiently accurate store aerodynamic load and trajectory prediction.