MDO Optimization of a Supersonic Aircraft by Reduced Order Models

In this paper a Fighter Supersonic Jet model is optimized to minimize RCS electromagnetic scattering and maximize aerodynamic efficiency. To reduce the computational effort while keeping at the same time a high level of reliability, Reduced Order Models (ROM) are defined to approximate the aerodynamic and the electromagnetic fields of the Jet, as a function of a reduced number of modes (principal components). The entire process of simulation automation and optimization has been integrated into the VOLTA web- based collaboration platform. Geometrical parameters related to the shape of the wings and tail planes are automatically updated on the CAD model, and CFD and electromagnetic simulations are executed on a distributed computational network using a Eulerian solver and a RCS code, on given conditions of Mach number and emitting frequency. An accurate surrogate ROM model trained on the database made by the CFD fields and the RCS curves is then used by the MDO optimization algorithm to evaluate the performance of the designs proposed, until the defined objectives are optimized.