Latent space subdivision: Stable and controllable time predictions for fluid flow

S. Wiewel, B. Kim, V. C. Azevedo, B. Solenthaler, N. Thuerey

Research output: Contribution to conferencePaperpeer-review

4 Scopus citations

Abstract

We propose an end-to-end trained neural network architecture to robustly predict the complex dynamics of fluid flows with high temporal stability. We focus on single-phase smoke simulations in 2D and 3D based on the incompressible Navier-Stokes (NS) equations, which are relevant for a wide range of practical problems. To achieve stable predictions for long-term flow sequences with linear execution times, a convolutional neural network (CNN) is trained for spatial compression in combination with a temporal prediction network that consists of stacked Long Short-Term Memory (LSTM) layers. Our core contribution is a novel latent space subdivision (LSS) to separate the respective input quantities into individual parts of the encoded latent space domain. As a result, this allows to distinctively alter the encoded quantities without interfering with the remaining latent space values and hence maximizes external control. By selectively overwriting parts of the predicted latent space points, our proposed method is capable to robustly predict long-term sequences of complex physics problems, like the flow of fluids. In addition, we highlight the benefits of a recurrent training on the latent space creation, which is performed by the spatial compression network. Furthermore, we thoroughly evaluate and discuss several different components of our method.

Original languageEnglish
Pages15-25
Number of pages11
DOIs
StatePublished - 6 Oct 2020
Event2020 ACM SIGGRAPH/Eurographics Symposium on Computer Animation, SCA 2020 - Virtual, Online, Canada
Duration: 6 Oct 20209 Oct 2020

Conference

Conference2020 ACM SIGGRAPH/Eurographics Symposium on Computer Animation, SCA 2020
Country/TerritoryCanada
CityVirtual, Online
Period6/10/209/10/20

Fingerprint

Dive into the research topics of 'Latent space subdivision: Stable and controllable time predictions for fluid flow'. Together they form a unique fingerprint.

Cite this