Innovating Works

IODA

Financiado
Industrial optimal design using adjoint CFD
Adjoint-based methods have become the most interesting approach in numerical optimisation using Computational Fluid Dynamics (CFD) due to their low computational cost compared to other approaches. The development of adjoint solve... Adjoint-based methods have become the most interesting approach in numerical optimisation using Computational Fluid Dynamics (CFD) due to their low computational cost compared to other approaches. The development of adjoint solvers has seen significant research interest, and a number of EC projects have been funded on adjoint-based optimisation. In particular, partners of this proposal are members of the EC FP7 projects FlowHead and AboutFlow which develops complete adjoint-based design methods for steady-state and unsteady flows in industrial design. Two related bottlenecks of applying goal-based optimisation in CFD are addressed here a) the efficient but flexible and automatic parametrisation of arbitrary shapes, and b) the imposition of design constraints. Parametrisation is at the core of optimisation, it defines the design space that the optimising algorithm is exploring. A range of parametrisations will be developed in the project, ranging from simple CAD-free methods with rich design spaces to CAD-based methods that return the optimised shape in CAD form. Integration of the currently available shape and topology modification approaches with the gradient-based optimisation approach will be addressed, in particular development of interfaces to return optimised CAD-free shapes into CAD for further design and analysis, an aspect that currently requires manual interpretation by an expert user. Constraints are at the core of industrial design, e.g. an optimised climate ducts for a vehicle needs to fit into the available build space. The project will develop efficient ways to extract constraints specified in the CAD model and apply them to CAD-free parametrisations. Methods will be developed to quantify how much the limited design space impairs the optimum and then to adaptively refine it. The results of the project will be applied to realistic mid-size and large-scale industrial optimisation problems supplied by the industrial project partners ranging from ver más
31/12/2018
QMUL
4M€
Duración del proyecto: 48 meses Fecha Inicio: 2014-12-05
Fecha Fin: 2018-12-31

Línea de financiación: concedida

El organismo H2020 notifico la concesión del proyecto el día 2018-12-31
Línea de financiación objetivo El proyecto se financió a través de la siguiente ayuda:
MSCA-ITN-2014-ETN: Marie Skłodowska-Curie Innovative Training Networks (ITN-ETN)
Cerrada hace 10 años
Presupuesto El presupuesto total del proyecto asciende a 4M€
Líder del proyecto
QUEEN MARY UNIVERSITY OF LONDON No se ha especificado una descripción o un objeto social para esta compañía.
Perfil tecnológico TRL 4-5