fPINN-DeepONet: A Physics-Informed Operator Learning Framework for Multi-Term Time-Fractional Mixed Diffusion–Wave Equations

We propose fPINN-DeepONet, a physics-informed operator-learning framework that approximates the solution operator of multi-term time-fractional mixed diffusion–wave equations. The architecture combines DeepONet’s branch–trunk decomposition with a fractional-PINN residual that handles nonlocal Caputo derivatives, generalizing across varying fractional orders, source terms, and boundary conditions in a single trained model.

Recommended citation: Lu, B., Hao, Z.-P., Moya, C., & Lin, G. (2025). "fPINN-DeepONet: A physics-informed operator learning framework for multi-term time-fractional mixed diffusion–wave equations." Journal of Computational Physics, 538, 114184.
Download Paper