A generic deep learning framework for propagation and scattering of acoustic waves in quiescent flows

Abstract

A deep learning surrogate for the direct numerical prediction of two-dimensional acoustic waves propagation and scattering with obstacles is developed through an auto-regressive spatiotemporal convolutional neural network. A single database of high-fidelity lattice Boltzmann temporal simulations is employed in the training of the network, achieving accurate predictions for long simulation times for a variety of test cases, representative of bounded and unbounded configurations. The capacity of the network to extrapolate outside the manifold of examples seen during the training phase is demonstrated by the obtaining of accurate acoustic predictions for relevant applications, such as the scattering of acoustic waves on an airfoil trailing edge, an engine nacelle or in-duct propagation. The method is tested for two types of input normalizations, coupled with an a-posteriori correction which improves the acoustic energy conservation of the predictions. The use of an adaptive local normalization along with the physics-based energy conservation results in an error reduction for all the studied cases.