This presentation was made at NAFEMS UK Conference 2018, "Taking Engineering Analysis and Simulation to the Next Level".
The NAFEMS UK Conference 2018 brought together all those involved in analysis and simulation from every corner of industry and academia, giving them an opportunity to advance their knowledge, give their organisations a competitive advantage, and a chance to be part of improving the technology itself.
An innovative computational approach, integrating mesh generation, CFD simultaneous analysis of noise source and propagation with acoustic radiation is presented and applied to the simulation of the Advanced Noise Control Fan (ANCF) developed by NASA Glenn Research Center.
The tonal noise source and the sound propagation in the nacelle duct and the nacelle near field are simultaneously predicted, starting from the engine geometry and parameters, with a single CFD analysis based on the non-linear harmonic (NLH) method. The sound radiation to the far field is then computed with the Green’s function approach implemented in a BEM frequency domain solver of the convective Helmholtz equation.
The present method provides a gain of close to two orders of magnitude compared to standard approaches, based on time-based full unsteady simulations. Instead of integrating in time the present approach uses a Fourier decomposition of the flow variables unsteady variations. Only selected frequencies are accounted for, which turns out as an advantage when focusing on clear tonal noise.
The nonlinear harmonic method is mathematically very interesting, as it has been shown that bringing the Fourier terms induces the appearance of additional systems of equations that have the same nature as the time-averaged ones. Those extra systems can be solved with the same numerical approach and same acceleration techniques, solving for the amplitudes and phases of the signal.
Finally it is shown that the presence of liners can be taken into account in a very simple way, by integrating impedance boundary conditions.
The setup and monitoring of harmonic runs is quite straightforward, as all equations just need to be converged to pseudo-steady state, which is much more convenient than standard unsteady runs. The computational cost of harmonic runs is also limited, as the overhead compared to steady state is very easily computed, based on the extra number of equations.