Physics-informed neural networks for one-dimensional sound field predictions with parameterized sources and impedance boundaries

Abstract

Realistic sound is essential in virtual environments, such as computer games and mixed reality. Efficient and accurate numerical methods for pre-calculating acoustics have been developed over the last decade; however, pre-calculating acoustics makes handling dynamic scenes with moving sources challenging, requiring intractable memory storage. A physics-informed neural network (PINN) method in 1D is presented, which learns a compact and efficient surrogate model with parameterized moving Gaussian sources and impedance boundaries, and satisfies a system of coupled equations. The model shows relative mean errors below 2%/0.2 dB and proposes a first step in developing PINNs for realistic 3D scenes.

11 pages, 5 figures, 3 tables

References in corpus (4)

• Implicit Neural Representations with Periodic Activation Functions
• Solving the wave equation with physics-informed deep learning
• Fast acoustic scattering using convolutional neural networks
• Dynamic Portal Occlusion for Precomputed Interactive Sound Propagation

Cited by in corpus (6)

• Data-aided Underwater Acoustic Ray Propagation Modeling
• Physics-informed neural network for acoustic resonance analysis in a one-dimensional acoustic tube
• Machine Learning in Acoustics: A Review and Open-Source Repository
• Point Neuron Learning: A New Physics-Informed Neural Network Architecture
• A convolutional plane wave model for sound field reconstruction
• Deep Room Impulse Response Completion