Fpinn-deeponet: An operator learning framework for multi-term time-fractional mixed diffusion-wave equations

Published in Journal of Computational Physics (2025)

We propose Fpinn-deeponet, a novel operator learning framework for solving complex time-fractional mixed diffusion-wave equations. This method leverages DeepONets integrated with physics-informed loss terms to accurately approximate solutions to nonlocal PDEs with varying fractional orders.

Authors: B Lu, ZP Hao, C Moya, G Lin
DOI: https://doi.org/10.1016/j.jcp.2025.114184