Sequence to sequence AE-ConvLSTM network for modelling the dynamics of PDE systems

2022/08/15 by Kakka, Priyesh Rajesh
#Computational Physics (physics.comp-ph) #FOS: Physical sciences #Fluid Dynamics (physics.flu-dyn)

paper · doi:10.48550/arxiv.2208.07315

Abstract

The article explains the convolutional LSTM (ConvLSTM) network in detail and introduces an improved auto-encoder version of the ConvLSTM network called AE-ConvLSTM. AE-ConvLSTM is also a sequence to sequence network that can predict long time evolution of a dynamical system by passing hidden states from one encoder to another. The network performed well in predicting the dynamic evolution of unsteady 2-D viscous Burgers when trained using data and, in another case, using governing equation (without data), i.e., physics-constrained. Further, AE-ConvLSTM was used in an effort to predict the time evolution of two unsteady Navier-Stokes problems. These problems have coupled pressure and velocity field having different magnitude order, and these fields evolve in time at a different rate. It was observed that the network could be trained using data, but while training using physics-constrained via governing equations, AE-ConvLSTM fails to train for time evolution.

Related

• FiniteNet: A Fully Convolutional LSTM Network Architecture for Time-Dependent Partial Differential Equations
• Fluid dynamics meet network science: two cases of temporal network eigendecomposition
• On the relationship between Koopman operator approximations and neural ordinary differential equations for data-driven time-evolution predictions
• PDE-Net: Learning PDEs from Data
• Latent-space Physics: Towards Learning the Temporal Evolution of Fluid Flow