Our research: aims and goals
FARE-X-AROMA-CFD project deals with advanced numerical analysis for parametric partial differential equations (PDEs) to improve scientific computing performances in more and more complex computational mechanics problems.
The aim of the FARE-X-AROMA-CFD project is to guarantee two further goals for ERC funded AROMA CFD project and to consolidate a strong critical mass in the development of numerical methods for computational reduction in fluid-dynamics and applications: computation reduction techniques, compressible flows and reduction in parametric space.
The first goal aims at the investigation of reduction strategies for the parameter space in order to be able to deal with problems with only a reasonable number of important parameters. This would help in the evaluation of sensitivity analysis studies, in dealing with more and more complex systems, properly parametrised with the important quantities. Parameter space reduction is a discipline with a growing importance and a big potential. Active subspaces, manifold learning and other proper techniques will be further investigated and adapted, especially to deal with different kind of parametrisation (geometry, physics, etc). This first part of proposed research needs to be properly integrated with computational reduction strategies developed in AROMA-CFD and FARE-X-AROMA-CFD. A versatile open source software tool will be properly developed, and planned to be used as preprocessing, but fully integrated in the computational pipeline.
The second goal aims at developing reduction strategies for the computational reduction of parametric aero-fluid dynamics compressible flows. This further topic allows to cover a full range of phenomena from aerodynamics to aero-acoustics, and of applications from aeronautics to turbo-machinery, as well as wind engineering, including multi-physics phenomena such as aero-thermo-elasticity. This plan is complimentary with AROMA-CFD where we deal with computational incompressible fluid dynamics problems in naval and mechanical engineering, as well as for cardiovascular flows. The proposed organisation of FARE-X-AROMA-CFD will follow the one scheduled for ERC AROMA-CFD, including strong methodological developments (stability, nonlinearities, inverse problems, flow control, uncertainty quantification, domain decomposition), then multi-physics applications, focused on aero-thermo-elasticity, as well as software developments: a new dedicated section for ITHACA -In (Real) Time Highly Advanced Computational Applications- open source library, and further subsections of ITHACA in the parts already under development with AROMA-CFD.
Several issues are expected to be further improved combining computational reduction strategies, like reduced order methods, with parameter space reduction, in view of more efficient (shape) optimisation and flow control problems. Not only, FARE-X-AROMA-CFD will allow to have a complete tool to face a wider range of fluid (aero) dynamics parametric problems and applications.