CAN-PINN: A fast physics-informed neural network based on coupled-automatic–numerical differentiation method

In this study, novel physics-informed neural network (PINN) methods for coupling neighboring support points and their derivative terms which are obtained by automatic differentiation (AD), are proposed to allow efficient training with improved accuracy. PINNs constrain their training loss function w...

Full description

Saved in:

Bibliographic Details
Published in:	Computer methods in applied mechanics and engineering Vol. 395; p. 114909
Main Authors:	Chiu, Pao-Hsiung, Wong, Jian Cheng, Ooi, Chinchun, Dao, My Ha, Ong, Yew-Soon
Format:	Journal Article
Language:	English
Published:	Amsterdam Elsevier B.V 15.05.2022 Elsevier BV
Subjects:	Accuracy Backward facing steps Channel flow Collocation Coupled-automatic–numerical differentiation Coupling Fluid flow Incompressible flow Interpolation Inverse problem Mathematical analysis Navier–Stokes equations Neural networks Numerical differentiation Operators (mathematics) Partial differential equations Physics Physics-informed neural network Robustness (mathematics) Taylor series expansions Training Training loss formulation Taylor series expansions Navier–Stokes equations Physics-informed neural network Training loss formulation Coupled-automatic–numerical differentiation Inverse problem
ISSN:	0045-7825
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Be the first to leave a comment!