ICCS 2023: Convolutional Recurrent Autoencoder for Molecular-Continuum Coupling

Piet Jarmatz, Sebastian Lerdo, Philipp Neumann

ML data set generation for molecular-continuum simulations with MaMiCo.

Molecular-continuum coupled flow simulations are used in many applications to build a bridge across spatial or temporal scales. Hence, they allow to investigate effects beyond flow scenarios modeled by any single-scale method alone, such as a discrete particle system or a partial differential equation solver. On the particle side of the coupling, often molecular dynamics (MD) is used to obtain trajectories based on pairwise molecule interaction potentials. However, since MD is computationally expensive and macroscopic flow quantities sampled from MD systems often highly fluctuate due to thermal noise, the applicability of molecular-continuum methods is limited. If machine learning (ML) methods can learn and predict MD based flow data, then this can be used as a noise filter or even to replace MD computations – both generates potential for tremendous speed-up of molecular-continuum simulations, aiming to enable new applications emerging on the horizon.

In this paper, we develop an advanced hybrid ML model for MD data in the context of coupled molecular-continuum flow simulations. Our model is based on recent know-how from computer vision and speech recognition, applied to a computational fluid dynamics context: A convolutional autoencoder (ConvAE) deals with the spatial extent of the flow data, while a recurrent neural network (RNN) is used to capture its temporal correlation. We use the open source coupling tool MaMiCo to generate MD datasets for ML training and implement the hybrid model as a PyTorch-based filtering module for MaMiCo.

The ML models are trained with real MD data from different flow scenarios including a Couette flow validation setup and a three-dimensional vortex street test case. Our results show that the convolutional recurrent hybrid model is able to learn and predict smooth flow quantities, even for very noisy MD input data. We furthermore demonstrate that also the more complex vortex street continuum flow data can accurately be reproduced by the ML module, without access to any corresponding continuum flow information.

This subdirectory contains a file structure of configurations in order to generate the MD datasets. In this paper, we use the macro-micro coupling tool (MaMiCo) to generate datasets based on Couette and Karman-Vortex-Street scenarios for molecular-continuum coupled flow simulations.

This subdirectory contains the single model approach as used for the Couette based datasets.

This subdirectory contains the triple model approach as used for the KVS based datasets.

This subdirectory contains our figures.

This subdirectory is a code graveyard.