Chair of Fluid Mechanics and Institute of Aerodynamics (AIA), RWTH Aachen University, Aachen (Germany)
Local Project ID:
HPC Platform used:
Hazel Hen of HLRS
Noise reduction is a key goal in European aircraft policy. One of the major noise sources at aircraft take-off is the engine jet noise. Recently, chevron nozzles were introduced which have drawn a lot of attention in research and the aircraft industry. Since the flow structures in the jet depend on the details of the nozzle exit geometry and have a large impact on the noise sources in the jet, scientists of the RWTH Aachen University extensively investigate chevron nozzles by running large-scale simulations based on a highly resolved mesh with up to 1 billion mesh cells.
Noise reduction is one of the major challenges of today’s aircraft development and together with an increase in fuel efficiency one of the key goals in European aircraft policy. The perceived noise levels of flying aircraft are to be reduced until 2050 by 65% compared to 2000. Various ideas exist to reduce aircraft noise. One of the major noise sources at take-off is the engine jet noise. Recently, serrated or chevron nozzles were introduced, since the flow structures in the jet depend on the details of the nozzle exit geometry and have a large impact on the noise sources in the jet. This kind of passive device has drawn a lot of attention in research and aircraft industry and is also investigated in this project. Figure 1 shows a model for a baseline configuration of an aircraft engine nozzle and a chevron nozzle.
A major problem of chevrons is, that while they can lead to a noise reduction during aircraft take-off, this nozzle design can, however, also lead to a severe loss of thrust during both take-off and cruise flight and thus to a lower efficiency of the engine. In addition, the noise reduction of chevrons can be intricate. They can reduce low frequency noise at large aft angles but at the same time increase high frequency noise at sideline angles.
These two aspects lead to a shape-optimization problem for the chevron nozzle design with the goal of reducing the low-frequency noise under the constraint of avoiding thrust reduction and an increase in high-frequency noise. For the shape optimization the influence of several design parameters, like the number of chevrons, penetration angles and the chevron length should be investigated.
High-performance computing (HPC) can help to tackle these research questions in a multi-stage process. In the first step, the unsteady compressible flow field of the jet is computed by using a so-called large-eddy simulation. In a second step, acoustic sources are determined from the flow field and acoustic perturbation equations are solved to determine the noise generation and propagation, i.e. the resulting acoustic sound field. With an optimization strategy, the nozzle design will be modified with the goal of noise reduction while maintaining the thrust of the engine.
The correct prediction of the jet noise, i.e. pressure waves of small amplitude, requires an accurate simulation of the flow field. This can only be achieved with simulations based on a highly resolved mesh with up to 1 billion mesh cells, since the fully turbulent flow field contains a broad range of scale lengths. To conduct such simulations, efficient, memory optimized and fully parallelized algorithms are required, that are adapted to high-performance computing hardware such as the Hazel Hen supercomputer installed at HLRS.
The effect of the chevron nozzles on the flow field can be seen in Figure 2. Here, the turbulent kinetic energy of two jets, exhausting from a circular baseline and a chevron nozzle, is visualized. As it can be seen in Figure 2, chevrons obviously enhance the mixing process. This weakens the formation of large-scale structures and excites smaller scale structures. Smaller turbulent scales are associated with high-frequency noise generation and large-scale structures to low-frequency noise.
Figure 3 shows contours of constant Mach number, coloured by the temperature of the flow field. The small-scale turbulent structures are clearly visible in the mixing layers of the jet.
Figure 4 shows sound waves that are generated in the jet flow and propagate into the far field, which shows the preferred sound propagation at an angle of about 30o – 40o relative to the axis of the main flow direction. In the next steps an optimization for the chevron geometry will be performed using the computational resources of HazelHen.
Matthias Meinke (PI), Vitali Pauz, Wolfgang Schröder
Chair of Fluid Mechanics and Institute of Aerodynamics, RWTH Aachen University
Chair of Fluid Mechanics and Institute of Aerodynamics
RWTH Aachen University
Wüllnerstr. 5a, D-52062 Aachen (Germany)
email: m.meinke [at] aia.rwth-aachen.de