Optimal Sparsity in Nonlinear Reduced Order Models Applied to an All-Movable Wing with Freeplay in Transonic Flow
Machine learning and artificial intelligence algorithms typically require large amount of data for training. This means that for nonlinear aeroelastic applications, where small training budgets are driven by the high computational burden associated with generating data, usability of such methods has been limited to highly simplified aeroelstic systems. In this paper, a sparsity promoting algorithm is used to significantly reduce the amount of training data required for the identification of the higher-order polynomial-based aeroelastic reduced order model (ROM). The study demonstrates that through orthogonal matching pursuit, it’s possible to efficiently identify optimized s-sparse nonlinear aerodynamic ROMs. This approach is exemplified in a three-dimensional aeroelastic stabilator model experiencing high amplitude freeplay-induced limit cycles. The comparison shows excellent agreement between the reduced order model and the full-order aeroelastic model. The development of an Optimally Sparse ROM (OS-ROM) highlights the feasibility of applying accurate, higher-order polynomial-based ROMs to complex nonlinear aeroelastic problems without incurring significant computational burdens.
