Physics-informed neural network for acoustic resonance analysis in a one-dimensional acoustic tube

Abstract

This study devised a physics-informed neural network (PINN) framework to solve the wave equation for acoustic resonance analysis. The proposed analytical model, ResoNet, minimizes the loss function for periodic solutions and conventional PINN loss functions, thereby effectively using the function approximation capability of neural networks while performing resonance analysis. Additionally, it can be easily applied to inverse problems. The resonance in a one-dimensional acoustic tube, and the effectiveness of the proposed method was validated through the forward and inverse analyses of the wave equation with energy-loss terms. In the forward analysis, the applicability of PINN to the resonance problem was evaluated via comparison with the finite-difference method. The inverse analysis, which included identifying the energy loss term in the wave equation and design optimization of the acoustic tube, was performed with good accuracy.

arXiv comment15 pages, 18 figures. This article may be downloaded for personal use only. Any other use requires prior permission of the author and AIP Publishing. This article appeared in [The Journal of the Acoustical Society of America, 156(1), 30-43 (2024)] and may be found at https://doi.org/10.1121/10.0026459

References in corpus (12)

• Adam: A Method for Stochastic Optimization
• Deep Learning in Neural Networks: An Overview
• DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators
• Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains
• Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks
• Physics-informed Neural Networks (PINNs) for Wave Propagation and Full Waveform Inversions
• Deep Fluids: A Generative Network for Parameterized Fluid Simulations
• Solving the wave equation with physics-informed deep learning
• Neural Network Models for the Anisotropic Reynolds Stress Tensor in Turbulent Channel Flow
• Neural Networks Fail to Learn Periodic Functions and How to Fix It
• Physics-informed neural networks for one-dimensional sound field predictions with parameterized sources and impedance boundaries
• Parallel Tacotron 2: A Non-Autoregressive Neural TTS Model with Differentiable Duration Modeling