An acoustic analogy is meant to calculate the acoustic pressure far from an object emitting sound (in the far field), precisely to alleviate the need for a computational domain and a grid. Then, you can cluster the computational effort in the noise source region.

In aeroacoustics, we are interested to find the noise at a far-field location (very far away from the body that produces the noise; say the noise that a human on the ground would hear from an airplane flying thousands of feet above him).

If we wanted to compute the noise directly using the fully compressible Navier-Stokes equation, then we would need a very fine grid with a very low time step upto the far-field {i.e., a huge domain which includes the airplane in air and the human on the ground} (which will usually be several tens of diameters away from the body)- this is computationally prohibitive.

That is where acoustic analogies, like FW-H, comes into play. Here we solve the near-field (just very close to the body of interest) using CFD and then use the data from that mesh (certain defined surfaces and volumes within the CFD mesh) to compute the terms in the FW-H integral, which would predict the acoustic pressure (noise) at any far-field location which is far away from the source of noise.

Here we assume one way coupling (only the aerodynamics affect the sound and not the other way), propagation in a homogenous medium, no reflection, etc. (there could be other assumptions which I may not be remembering now).

Good topic

An acoustic analogy is meant to calculate the acoustic pressure far from an object emitting sound (in the far field), precisely to alleviate the need for a computational domain and a grid. Then, you can cluster the computational effort in the noise source region.

can i use FW-H for near field acoustic calculations? Ronith Stanly Stéphane Moreau