Version 1
: Received: 18 June 2020 / Approved: 21 June 2020 / Online: 21 June 2020 (10:56:06 CEST)
Version 2
: Received: 30 September 2020 / Approved: 30 September 2020 / Online: 30 September 2020 (03:51:25 CEST)
How to cite:
Zhang, Q.; Chen, Y.; Yang, Z. Data-Driven Solutions and Discoveries in Mechanics Using Physics Informed Neural Network. Preprints2020, 2020060258. https://doi.org/10.20944/preprints202006.0258.v2
Zhang, Q.; Chen, Y.; Yang, Z. Data-Driven Solutions and Discoveries in Mechanics Using Physics Informed Neural Network. Preprints 2020, 2020060258. https://doi.org/10.20944/preprints202006.0258.v2
Zhang, Q.; Chen, Y.; Yang, Z. Data-Driven Solutions and Discoveries in Mechanics Using Physics Informed Neural Network. Preprints2020, 2020060258. https://doi.org/10.20944/preprints202006.0258.v2
APA Style
Zhang, Q., Chen, Y., & Yang, Z. (2020). Data-Driven Solutions and Discoveries in Mechanics Using Physics Informed Neural Network. Preprints. https://doi.org/10.20944/preprints202006.0258.v2
Chicago/Turabian Style
Zhang, Q., Yilin Chen and Ziyi Yang. 2020 "Data-Driven Solutions and Discoveries in Mechanics Using Physics Informed Neural Network" Preprints. https://doi.org/10.20944/preprints202006.0258.v2
Abstract
Deep learning has achieved remarkable success in diverse computer science applications, however, its use in other traditional engineering fields has emerged only recently. In this project, we solved several mechanics problems governed by differential equations, using physics informed neural networks (PINN). The PINN embeds the differential equations into the loss of the neural network using automatic differentiation. We present our developments in the context of solving two main classes of problems: data-driven solutions and data-driven discoveries, and we compare the results with either analytical solutions or numerical solutions using the finite element method. The remarkable achievements of the PINN model shown in this report suggest the bright prospect of the physics-informed surrogate models that are fully differentiable with respect to all input coordinates and free parameters. More broadly, this study shows that PINN provides an attractive alternative to solve traditional engineering problems.
Keywords
Conservation laws; Data inference; Data discovery; Dimensionless form; PINN
Subject
Engineering, Civil Engineering
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Commenter: Qi Zhang
Commenter's Conflict of Interests: Author