Nilufer K. Bulut
This study demonstrates that hybrid training strategies can enhance the accuracy and energy consistency of Physics-Informed Neural Networks (PINNs) for electromagnetic wave propagation, making them competitive with traditional methods like FDTD.
Physics-Informed Neural Networks (PINNs) are a new way of solving problems in physics by incorporating the laws of physics directly into the training of neural networks. In the field of electromagnetism, traditional methods like FDTD are well-established, so new methods must offer significant benefits to be adopted. This research shows that by using hybrid training techniques, PINNs can improve their accuracy and energy consistency to match the performance of FDTD. The study uses innovative strategies to address common issues in wave propagation, making PINNs a promising alternative for solving electromagnetic problems.
