Auto-Encoded Reservoir Computing for Turbulence Learning

TL;DR

This paper introduces an Auto-Encoded Reservoir Computing method that effectively learns and predicts the complex dynamics of 2D turbulent flows by combining manifold learning with time evolution modeling.

Contribution

It presents a novel combination of autoencoders and reservoir computing to improve turbulence modeling and prediction capabilities.

Findings

Successfully learns flow dynamics and predicts statistical moments.

Achieves accurate spatio-temporal turbulence predictions.

Demonstrates potential for machine learning in turbulence research.

Abstract

We present an Auto-Encoded Reservoir-Computing (AE-RC) approach to learn the dynamics of a 2D turbulent flow. The AE-RC consists of an Autoencoder, which discovers an efficient manifold representation of the flow state, and an Echo State Network, which learns the time evolution of the flow in the manifold. The AE-RC is able to both learn the time-accurate dynamics of the flow and predict its first-order statistical moments. The AE-RC approach opens up new possibilities for the spatio-temporal prediction of turbulence with machine learning.