NASA aims to improve the sonic environment
A key challenge in aircraft design is to minimize the noise radiated during take-off and landing. Increasingly strict U.S. regulations on noise pollution surrounding airports near major metropolitan areas mandate that aircraft that do not meet noise requirements be retrofitted or removed from service. Ultimately, quieter designs mean improvements to the sonic environment for communities near airports, greater flexibility for the airlines, and increased profits for aircraft manufacturers.
Aerodynamic sound is generated by the unsteady motion of a gas and its interaction with surrounding surfaces. These interactions produce minute disturbances in the local pressure field (waves) that propagate outward from the source. It is well-known that, in addition to engine noise, sound generated by the airframe constitutes a major component of total aircraft noise during approach and landing. NASA, in partnership with Gulfstream Aerospace, is studying the causes and mitigation strategies for airframe noise. Previous studies have found that a significant portion of airframe noise can be attributed to the landing gear. The work conducted under the NASA-Gulfstream partnership is partly experimental (flight tests and wind-tunnel studies) and partly computational in nature. The latter effort is our focus here.
CFD and high performance computing
Since its infancy during the 1960s, the discipline of computational fluid dynamics (CFD), which simulates the interactions between a fluid flow and structures of arbitrary shape, has continually contributed to the advancement of high performance computing (HPC). Growth in computational capability has made airframe noise problems deemed to be near impossible only 15 years ago to be now within the realm of what can be achieved in a cost-effective manner and reasonable time. Although the high-fidelity simulation of airframe noise generated by a full aircraft in landing configuration is still beyond the capabilities of even the most powerful supercomputers, CFD simulations of aircraft components, such as a nose landing gear, are within our reach and are being undertaken by several independent groups within the airframe noise community.
Scientists at NASA Langley Research Center in Hampton, VA, are utilizing the internally developed CFD solver FUN3D to simulate the complex unsteady flow surrounding a Gulfstream G550 aircraft nose landing gear. Such simulations are essential to
- provide insight on the nature of noise sources
- guide physics-based noise source modeling
- advance airframe noise prediction capabilities
- develop and evaluate noise reduction concepts
What makes the simulation of airframe noise in general — and of landing gear noise in particular — a challenging task are the intricate geometries involved and the broadband nature of the noise produced by these components. Close proximity of the many blunt bodies of various sizes and shapes that collectively make up a landing gear system produces an extremely complex flow field that is highly interactive and nonlinear. This field contains noise-generating flow structures (vortices) that cover a broad range of spatial and temporal scales.
Figure 1: Nose landing gear surface grid, including a partial view of the wind tunnel walls
An accurate characterization of aerodynamic noise requires proper spatial and temporal resolution of the flow structures. Proper spatial resolution demands that the volume mesh surrounding the landing gear be very fine (Figure 1). For the present G550 nose landing gear simulations, the volume grid contains 400 million cells.
To resolve and/or recover the high-frequency component of the landing gear noise spectrum, very small time steps must be used during the time-advancement cycle of the simulations; to recover the low frequency component of the noise spectrum, the simulations must be run for tens of thousands of time steps. At each time step, the FUN3D solver must execute 680 billion floating-point operations per second to advance the flow simulation on the 400 million cells to the next time step. To obtain a converged solution within a reasonable amount of time, this vast number of operations must be executed using highly efficient parallel algorithms on large supercomputers.
The present high-fidelity nose gear computation was performed on NASA’s Pleiades supercluster, located at Ames Research Center, Moffett Field, CA. The simulation used 1,200 cores and took nearly two months of continuous running time to complete. Currently, research is underway in NASA Langley’s Computational AeroSciences Branch to extend the parallel processing capabilities of FUN3D to allow even larger computations — on grids with several billion cells using tens of thousands of processors — in order to reduce the solution turnaround time to just a few days, if not hours.
Visualization and data analysis challenges
An equally challenging aspect of the landing gear simulation is the analysis and visualization of the nose gear flow field. Proper visualization of the resulting flow interactions among the various gear components is critical to an overall understanding of the air flow and noise propagation surrounding the landing gear. We are using the current high-fidelity simulations to improve the state-of-the-art knowledge on several important aspects of the aeroacoustics field, such as determination of the type of noise being generated (broadband, tonal or a mixture of both); identification of the locations and strengths of the salient noise sources; establishment of best practices on numerical resolution (spatial and temporal) required to properly capture the key noise-producing flow features; creation of superior physics-based models of landing gear noise; and development of effective noise reduction technologies for application to current and future generations of civil transports.
The three sample animations included in this article illustrate the wealth of information that a researcher can glean from such visualizations. The first animation, displaying the unsteady vorticity field surrounding the nose gear, shows where the flow structures are created and how they interact with the gear subcomponents. The second captures the pressure fluctuations on the surface of the landing gear and shows the regions of high amplitude (hot spots), where most of the noise is generated. The third animation depicts the acoustic waves that result from the surface pressure fluctuations, as they radiate to the ground.
To generate the flow animations presented here required saving a small portion (12,000 snapshots or time steps) of the flow simulation record. With each snapshot resulting in a file size on the order of 4 to 5 gigabytes, the total time record saved is in excess of 50 to 70 terabytes of data. Although such an aggregated file size is not excessively large by today’s standards, it is still too large for routine visualization of the results. The push toward much larger simulations (a nose gear computation on a grid twice as large as the current grid is ongoing) precludes relying on traditional methods for post-processing of CFD data; that is, saving the volumetric information at each time step for analysis at a later time, as these are highly inefficient and no longer practical. Such large datasets demand concurrent real-time simulation, analysis and visualization of the flow field without the need to save countless terabytes of information that would soon tax the storage capacity of even the largest supercomputers.
Scientific visualization of high-fidelity, large-scale flow simulations such as these has become an indispensable tool for providing global insights and knowledge that enable the development of viable engineering solutions to pressing environmental issues affecting the public good. The landing gear simulations, for example, together with those from other disciplines relevant to aircraft design, will soon be used to help develop a new breed of subsonic aircraft that will not only reduce noise pollution, but will burn less fuel and produce fewer harmful emissions — all to improve life on our planet.
Mehdi Khorrami is lead project investigator and an aerospace technologist in the Computational AeroSciences Branch at NASA Langley Research Center, and Patrick Moran is a research scientist in the NASA Advanced Supercomputing (NAS) Division at NASA Ames Research Center. They may be reached at editor@ScientificComputing.com.