Fluid dynamics simulations are a valuable tool for improving the design of aircraft engines and rendering them more efficient. Now, a team of researchers has built the first high-fidelity simulation of a full aircraft engine including several coupled engine components — a milestone that provides insights into the interactions between the individual components and will help to improve future simulations.
Airplane engines work, no question about it. But their complex machineries could still be optimised for more efficiency. After all, more efficient engines require less fuel, which reduces cost as well as pollution of the atmosphere. That’s why engineers and scientists are continuously working on improving aircraft engine designs. In this process, computer simulations play an essential role since they expand and enhance experimental results by adding information that can hardly be gained from experimental setups. However, engine simulations that make use of supercomputers have been rather limited as well due to the vast computing power required for detailed and accurate simulations. That’s why, so far, simulations have mostly consisted of only one engine component, such as the high-pressure compressor or the combustion chamber. But now, Dr Carlos Pérez Arroyo and project leader Dr Jérôme Dombard together with their co-workers at CERFACS, a private laboratory in Toulouse, have successfully performed one of the first ever full-engine simulations.
Executed on PRACE supercomputers, the simulation included three engine components: the fan, the high-pressure compressor, and the combustion chamber. The project was done in close collaboration with the two aviation companies Safran SA, a large producer of helicopter and airplane engines, and Akira technology, a pioneering aircraft engine company that constructed the engine model DGEN380 that was simulated in this project, a so-called turbofan designed for business jets. Although the analysis is still in progress, the first results already demonstrate that such a multi-component simulation is feasible — a fact that was not self-evident before — and that it delivers significant benefits.
The full-engine simulation reveals the interactions between the different engine components. For instance, the incoming pressure wave from the compressor is detectable in the combustion chamber, here made visible as red and green pattern on the sides of the chamber.
Image credit: CERFACS / AKIRA / Project FULLEST
The benefit of coupling components
“When you couple different engine components, simulations become a lot more extensive and complex”, Dombard points out. To carry out the simulation, Dombard’s team member and postdoctoral researcher Carlos Pérez Arroyo used the AVBP code developed at CERFACS. His first challenge was to perform each of the single component simulations, which exhibit a variety of different physical processes. Secondly, he developed a methodology to couple the components and thus achieve the fully integrated simulation. Within the new setup, the so-called unsteady coupling between the different components is mathematically solved and corresponds to the most realistic and high-fidelity representation of the underlying physical processes.
Finally, for the representation of a full engine, it was necessary to build a 360-degrees simulation. In contrast, conventional simulations representing only one single component can benefit from the presence of repetitive elements — the blades of the fan for instance or the injectors of the combustion chamber. “In fact, many simulations of single components compute just one of these repetitive sections”, says Dombard. But when simulating several coupled components that each possess different architectures, this is impossible. Instead, the simulation has to cover the whole 360 degrees of the object. This increases the sheer size and complexity of the simulation drastically. In the end, the full 360-degree large-eddy simulation highly resolved the timeline of the combustion process, as well as the spatial resolution of the engine components represented by two billion cells.
The simulation included three adjacent components of the DGEN380 engine: the fan, the high-pressure compressor and the combustion chamber. Only within such a coupled representation can the physical interactions between the different components be investigated.
Image credit: CERFACS / AKIRA / Project FULLEST
Providing insights for the community
The outcome of the simulation is well worth the effort, because now, for the first time, the scientists can investigate the interactions between the different components and how they influence the processes within the engine. For example, the preliminary results revealed that there is a strong interaction between the high-pressure compressor and the combustion chamber. Specifically, the compressor generates a pressure wave with a very high amplitude that enters into the combustion chamber. This pulse alters the environment in the chamber and influences the whole mass flow, meaning the air and fuel flows taking part in the combustion process. “So far, no engine simulation took this effect into account”, says Pérez Arroyo. “Now, we have shown that coupling the components indeed makes a difference.” Moreover, preliminary results indicate that the pressure wave generated from the compressor also affects the upstream component, the fan.
The team will continue to analyse these interactions between components — not only between adjacent components such as the fan and compressor, but also between non-neighbouring components like the fan and the combustion chamber. Furthermore, they are already planning the next extension of the simulation: the incorporation of the last missing engine component, the turbine. Ultimately, the scientists aim to provide a complete high-fidelity database of unsteady coupled engine components to help other teams develop new simulations and validate existing ones.
Project title: FULLEST – First fUlL engine computation with Large Eddy SimulaTion
Resources awarded: 20th call: 31 600 000 core hours on Joliot-Curie (SKL) hosted by GENCI at CEA, France.
Research field: Engineering
Team: C. Pérez Arroyo, F. Duchaine, L. Gicquel, P. Mohanamuraly, N. Odier and G. Staffelbach, J. Dombard; technical support by N. Buffaz, G. Exilard and S. Richard (Safran Helicopter Engines), T. Quirante and N. Vieira-Nobre (AKIRA Technology).
Acknowledgements: This project was funded by the French agency DGAC and Safran Tech within the project ATOM. This work benefitted from collaboration with the EXCELLERAT project which has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under grant agreement 823691, as well as from the European Union’s Horizon 2020 Research and Innovation Programme via the EPEEC project, grant agreement 801051.