A nonlocal physics-informed deep learning framework using the peridynamic differential operator

https://www.researchgate.net/publication/341816480_A_nonlocal_physics-i…

Dear iMechanica Community: 

I am sharing with you our latest work on Physics-Informed Neural Networks (PINN) for the discovery of non-smooth problems with steep gradients (concentration) or singularity. To this end, we propose a nonlocal PINN architecture, inspired by the Peridynamic nonlocal theory, using the Peridynamic Differential Operator. 

If you are interested in deep learning for non-smooth problems with discontinuity, steep gradients, or even singularity, please check our pre-print at: 

https://www.researchgate.net/publication/341816480_A_nonlocal_physics-i…

Thanks,

Ehsan