Efficiency Enhancement of Aeroelastic Optimization Process Using Parametric Reduced-Order Modeling

In this work, to demonstrate the efficiency of model reduction in design optimization, a parametric reduced-order model (PROM) was adopted in conjunction with an aeroelastic optimization process. Flutter speed was chosen as an objective function, and structural properties (material density, Young's modulus, and Poisson's ratio) as well as fluid properties (air density) were defined as the design variables. The flutter calculation was performed for a Goland wing, using finite-element modeling for the structure and the vortex lattice method for the aerodynamic part. A gradient-based optimization technique and a global optimization method were used to seek a maximum flutter speed. Comparison of optimization results from the full-order model (FOM) and PROM shows that the proposed optimization process yields the same optimum flutter speed as the FOM and yet reduces the computation time significantly, by up to a factor of four. © 2018 American Society of Civil Engineers.