Final Report Summary - DESIDER (Detached Eddy Simulation for Industrial Aerodynamics)
The major aim of the DESIDER project was to overcome known weaknesses in the different turbulence-resolving approaches (detached-eddy simulation or DES, other RANS-LES hybrids, and scale adaptive simulation or SAS) in order to support the European aeronautics industry with simulation methods that offer increased predictive accuracy for complex turbulent flows.
The main objectives, that were achieved by gathering expertise from European experts in the field of aerodynamics, turbulence modelling and numerical analysis up to multi-disciplinary (aero-elastic and aero-acoustic) design, were as follows.
1. Based on the previously developed DES approach, advanced modelling approaches were investigated and developed for unsteady flow simulations as a compromise between URANS and LES, which are now able to produce LES-comparable results for real aeronautical applications, yet with less costly computational resources than using LES for an employment in industrial design environments.
2. The project demonstrated capabilities of hybrid RANS-LES approaches in solving industrially relevant applications with a focus on aerodynamic flows characterised by separation, wakes, vortex interaction and buffeting, i.e. all flows which are inherently unsteady.
3. It was further investigated that hybrid RANS-LES methods can be well applied to multi-disciplinary topics as there are aero-acoustics (noise reduction) and aero-elastics (reduced weight, unsteady loads, fatigue issues, improved safety), improving by this a cost-effective design.
4. The DESIDER project facilitated co-operation between European industries, research establishments and universities, it fostered co-operation, improve dissemination, and achieved cross-fertilisation between different industries as there are airframe, turbo-machinery, helicopters, power generation as well as turbo-engines and ground transportation by setting up a so-called 'observer group', i.e. a group of industries plus one additional university (Liverpool University) who were linked to the project but not paid by it. Some of the observers provided their own results as a gift in return for being able to attend the technical meetings of DESIDER.
Major studies performed in the course of DESIDER in terms of development and assessment of the turbulence-resolving approaches, include:
- investigation of a role of a background RANS model in DES in terms of accuracy and robustness;
- improvement of RANS-LES switching in DES and, particularly, elimination of premature switching which may occur inside the boundary layer with a grid that is not sufficient for the well-resolved LES;
- extension of SAS modelling to two-equation turbulence models and a thorough assessment of the SAS approach against standard URANS and DES methods;
- development of DES-based approaches for the near-wall treatment in LES which would enable DES application to flows without any separation zone;
- assessment of DES, RANS-LES hybrids, and SAS capabilities in aero-acoustics and aero-elastics analysis, i.e. proof-of-applicability of new models in multi-disciplinary design environments;
- development of a new experimental data base for channel flow, with measurements carried out by ONERA and data post-processing by University Lille.
Results have been used to test the hybrid RANS-LES methods investigated in the DESIDER project. Furthermore, as added value (i.e. not funded totally by DESIDER), cylinder measurements in the critical regime were performed by IMFT. Data of the latter experiment have also been used and validated in the DESIDER project.
All the findings made in the course of the above studies are supported by application of the different approaches to a wide range of thoroughly selected generic and industrial test cases. Based on this, the models limitations were established, best practice recommendations were formulated and, most important, the predictive capabilities of the in-house CFD codes of the partners, and in particularly industry, were significantly enhanced.
Moreover, the DESIDER project was making Europe and, hence, the European research community in the area of aeronautics, a world-leading group on the improvement and application of hybrid RANS-LES methods. Having pointed out that the results are of major concern for the aeronautics industry with respect to the improved predictive accuracy of CFD results, it should be of high interest not spoil that leadership. This strongly indicates to continue - at a European level - with activities on flow-physics modelling.
The main objectives, that were achieved by gathering expertise from European experts in the field of aerodynamics, turbulence modelling and numerical analysis up to multi-disciplinary (aero-elastic and aero-acoustic) design, were as follows.
1. Based on the previously developed DES approach, advanced modelling approaches were investigated and developed for unsteady flow simulations as a compromise between URANS and LES, which are now able to produce LES-comparable results for real aeronautical applications, yet with less costly computational resources than using LES for an employment in industrial design environments.
2. The project demonstrated capabilities of hybrid RANS-LES approaches in solving industrially relevant applications with a focus on aerodynamic flows characterised by separation, wakes, vortex interaction and buffeting, i.e. all flows which are inherently unsteady.
3. It was further investigated that hybrid RANS-LES methods can be well applied to multi-disciplinary topics as there are aero-acoustics (noise reduction) and aero-elastics (reduced weight, unsteady loads, fatigue issues, improved safety), improving by this a cost-effective design.
4. The DESIDER project facilitated co-operation between European industries, research establishments and universities, it fostered co-operation, improve dissemination, and achieved cross-fertilisation between different industries as there are airframe, turbo-machinery, helicopters, power generation as well as turbo-engines and ground transportation by setting up a so-called 'observer group', i.e. a group of industries plus one additional university (Liverpool University) who were linked to the project but not paid by it. Some of the observers provided their own results as a gift in return for being able to attend the technical meetings of DESIDER.
Major studies performed in the course of DESIDER in terms of development and assessment of the turbulence-resolving approaches, include:
- investigation of a role of a background RANS model in DES in terms of accuracy and robustness;
- improvement of RANS-LES switching in DES and, particularly, elimination of premature switching which may occur inside the boundary layer with a grid that is not sufficient for the well-resolved LES;
- extension of SAS modelling to two-equation turbulence models and a thorough assessment of the SAS approach against standard URANS and DES methods;
- development of DES-based approaches for the near-wall treatment in LES which would enable DES application to flows without any separation zone;
- assessment of DES, RANS-LES hybrids, and SAS capabilities in aero-acoustics and aero-elastics analysis, i.e. proof-of-applicability of new models in multi-disciplinary design environments;
- development of a new experimental data base for channel flow, with measurements carried out by ONERA and data post-processing by University Lille.
Results have been used to test the hybrid RANS-LES methods investigated in the DESIDER project. Furthermore, as added value (i.e. not funded totally by DESIDER), cylinder measurements in the critical regime were performed by IMFT. Data of the latter experiment have also been used and validated in the DESIDER project.
All the findings made in the course of the above studies are supported by application of the different approaches to a wide range of thoroughly selected generic and industrial test cases. Based on this, the models limitations were established, best practice recommendations were formulated and, most important, the predictive capabilities of the in-house CFD codes of the partners, and in particularly industry, were significantly enhanced.
Moreover, the DESIDER project was making Europe and, hence, the European research community in the area of aeronautics, a world-leading group on the improvement and application of hybrid RANS-LES methods. Having pointed out that the results are of major concern for the aeronautics industry with respect to the improved predictive accuracy of CFD results, it should be of high interest not spoil that leadership. This strongly indicates to continue - at a European level - with activities on flow-physics modelling.