The ability of a weakly compressible solver to solve landing gear noise with flow-acoustic coupling is investigated. Traditionally compressible flow solvers are used to simulate flow-acoustic coupling. However, compressible solvers require greater computational resources compared to incompressible solvers at low Mach numbers such as an aircraft on approach to landing. While incompressible solvers are able to capture the steady aerodynamics, in this work, their inability to capture the correct wall pressure spectra and far-field acoustics is demonstrated for a relatively simple two wheel landing gear, i.e. the LAGOON (Landing Gear Noise database for CAA validation) #1 geometry. This is due to the inability of incompressible solvers to capture cavity resonances, which are important contributors to not only tonal but also broadband noise. This is significant for landing gear noise predictions. A weakly compressible solver is tested to determine its ability to resolve the flow-acoustic coupling for a low Mach number flow. This solver has a similar computational cost to an incompressible solver. The simulations are performed using a weakly compressible solver added to OpenFOAM with a one-equation Large-Eddy Simulation model. An incompressible simulation of the same configuration is also performed for comparison. Results shows that the weakly compressible solver can correctly solve the resonant tonal and broadband noise that are absent in the solution from the incompressible solver. The weakly compressible solver maintains similar behaviour in predicting the time-averaged and root-mean-square flow variables. A grid sensitivity study is also included, and consistency is compared between the two meshes of different resolutions for both near-field pressure fluctuations and far-field acoustics. The computational cost of the weakly compressible method for landing gear simulations is slightly less than the incompressible solver and significantly less than other fully compressible Navier-Stokes solvers.
展开▼
