In this paper the solution of industrial aerodynamic shape optimization problems using the optimization methods provided by the RBF4AERO platform, developed in the framework of the EU–funded RBF4AERO project, is demostred.

The platform provides a complete infrastructure needed for optimization problems, including GUI, Optimization Algorithms, CFD Solvers, Morpher Tool and Benchmark Management System. Both stochastic and gradient-based optimization methods are implemented on this platform.

The design variables are related to a morphing tool based on Radial Basis Functions (RBFs) which control the deformation of both surface and volume meshes. Whenever the optimization is based on gradient-based techniques, the continuous adjoint method is used to compute the sensitivity derivatives while the Morpher tools give the mesh deformation velocity.

The stochastic tool is based on Evolutionary Algorithms (EAs) assisted by surrogate evaluation models (Response Surface Methods, RSM). A sampling technique (Design of Experiments) provides the training patterns of the RSM which is exclusively used as the evaluation tool within the EA-based optimization. At the end of each EA-based optimization, the resulting “optimal” solution(s) are re-evaluated by means of the CFD tool, before proceeding to the next cycle if needed.

The optimization of car and aircraft models on the RBF4AERO platform is showcased.

The paper can be download in Presentations and Proceedings 2016.