PINNs Model Wave Propagation in Bimaterial Systems

Researchers have developed a Physics-Informed Neural Network (PINN) framework to model transient elastodynamic wave propagation within systems composed of two different materials. This innovative approach directly embeds the fundamental physical laws governing linear elasticity and wave dynamics into the neural network's learning process through a specialized loss function. The framework was validated using a steel-aluminum specimen, demonstrating its ability to accurately predict wave transmission, reflection, and displacement histories, closely matching high-fidelity finite-element simulations. The trained PINN acts as a continuous surrogate model, capable of predicting wave responses for unseen time points or modified material properties without requiring new simulations, offering a computationally efficient alternative for complex engineering analyses.