Efficient computation of the POD manifold containing the information required to generate a multi-parameter aerodynamic database

A method is presented to calculate efficiently the proper orthogonal decomposition (POD) manifold (spanned by POD modes) that contains the information needed to generate a multi-parameter aerodynamic database. This work is motivated by the need, driven by market competition, to improve the efficiency of both aerodynamic design and analysis processes in the aeronautic sector. Current aerodynamic database generation is carried out in industry performing CFD computations at the nodes of a dense Cartesian grid in the parameter space. This approach does not account for the redundancies that are necessarily involved and, therefore, it is not the most efficient from the points of view of computational cost and time needed to generate the database. Such redundancies are an advantage when generating the database by means of the (already described in the literature) reduced order models based on projection of the governing equations on a POD manifold, which must be previously calculated from a set of CFD computed snapshots. The method proposed in this paper provides such POD manifold with a small number of snapshots, comparable to the dimension of the POD manifold itself, which is a crucial step for the computational efficiency of the proposed database generation method.The method is tested considering the aerodynamic flow around a horizontal tail plane (HTP) in the parameter range 0.4⩽Mach number⩽0.80.4⩽Mach number⩽0.8 and −3°⩽angle of attack⩽3°−3°⩽angle of attack⩽3°, obtaining the POD manifold with a good precision using 20, conveniently selected snapshots.