Predicting parametric spatiotemporal dynamics by multi-resolution PDE structure-preserved deep learning

05/09/2022
by   Xin-Yang Liu, et al.
59

Although recent advances in deep learning (DL) have shown a great promise for learning physics exhibiting complex spatiotemporal dynamics, the high training cost, unsatisfying extrapolability for long-term predictions, and poor generalizability in out-of-sample regimes significantly limit their applications in science/engineering problems. A more promising way is to leverage available physical prior and domain knowledge to develop scientific DL models, known as physics-informed deep learning (PiDL). In most existing PiDL frameworks, e.g., physics-informed neural networks, the physics prior is mainly utilized to regularize neural network training by incorporating governing equations into the loss function in a soft manner. In this work, we propose a new direction to leverage physics prior knowledge by baking the mathematical structures of governing equations into the neural network architecture design. In particular, we develop a novel PDE-preserved neural network (PPNN) for rapidly predicting parametric spatiotemporal dynamics, given the governing PDEs are (partially) known. The discretized PDE structures are preserved in PPNN as convolutional residual network (ConvResNet) blocks, which are formulated in a multi-resolution setting. This physics-inspired learning architecture design endows PPNN with excellent generalizability and long-term prediction accuracy compared to the state-of-the-art black-box ConvResNet baseline. The effectiveness and merit of the proposed methods have been demonstrated over a handful of spatiotemporal dynamical systems governed by unsteady PDEs, including reaction-diffusion, Burgers', and Navier-Stokes equations.

READ FULL TEXT

page 13

page 16

page 20

page 21

page 23

page 25

page 27

page 31

research
04/14/2020

PhICNet: Physics-Incorporated Convolutional Recurrent Neural Networks for Modeling Dynamical Systems

Dynamical systems involving partial differential equations (PDEs) and or...
research
02/03/2020

PDE-NetGen 1.0: from symbolic PDE representations of physical processes to trainable neural network representations

Bridging physics and deep learning is a topical challenge. While deep le...
research
08/11/2023

PDE Discovery for Soft Sensors Using Coupled Physics-Informed Neural Network with Akaike's Information Criterion

Soft sensors have been extensively used to monitor key variables using e...
research
12/29/2018

Autoencoder Based Residual Deep Networks for Robust Regression Prediction and Spatiotemporal Estimation

To have a superior generalization, a deep learning neural network often ...
research
01/15/2020

Time evolution of the characteristic and probability density function of diffusion processes via neural networks

We investigate the use of physics-informed neural networks-based solutio...
research
01/27/2023

Improving deep learning precipitation nowcasting by using prior knowledge

Deep learning methods dominate short-term high-resolution precipitation ...

Please sign up or login with your details

Forgot password? Click here to reset