Generalizability of reservoir computing for flux-driven two-dimensional convection

Abstract

We explore the generalization properties of an echo state network applied as a reduced dynamical model to predict flux-driven two-dimensional turbulent convection. To this end, we consider a convection domain at fixed height with a variable ratio of buoyancy fluxes at the top and bottom boundaries, which break the top-down symmetry in comparison to the standard Rayleigh-Bénard case thus leading to highly asymmetric mean and fluctuation profiles across the layer. Our direct numerical simulation model describes a convective boundary layer in a simple way. The data are used to train and test a recurrent neural network in the form of an echo state network. The input to the echo state networks is obtained in two different ways, either by a proper orthogonal decomposition or by a convolutional autoencoder. In both cases, the echo state network reproduces the turbulence dynamics and the statistical properties of the buoyancy flux, and is able to model unseen data records with different flux ratios.

References in corpus (21)

• Machine Learning for Fluid Mechanics
• Turbulence Modeling in the Age of Data
• Using Machine Learning to Replicate Chaotic Attractors and Calculate Lyapunov Exponents from Data
• Deep Neural Networks for Data-Driven Turbulence Models
• Hybrid Forecasting of Chaotic Processes: Using Machine Learning in Conjunction with a Knowledge-Based Model
• Reduced-order modeling of advection-dominated systems with recurrent neural networks and convolutional autoencoders
• Machine learning based spatio-temporal super resolution reconstruction of turbulent flows
• Deep Neural Networks for Nonlinear Model Order Reduction of Unsteady Flows
• Machine-learning-based reduced order modeling for unsteady flows around bluff bodies of various shapes
• A Perspective on Machine Learning Methods in Turbulence Modelling
• Use of neural networks for stable, accurate and physically consistent parameterization of subgrid atmospheric processes with good performance at reduced precision
• Deep learning to discover and predict dynamics on an inertial manifold
• Controlling Rayleigh-Bénard convection via Reinforcement Learning
• Reconstruction of turbulent data with deep generative models for semantic inpainting from TURB-Rot database
• Reservoir computing model of two-dimensional turbulent convection
• Combining Machine Learning with Knowledge-Based Modeling for Scalable Forecasting and Subgrid-Scale Closure of Large, Complex, Spatiotemporal Systems
• Transfer learning for nonlinear dynamics and its application to fluid turbulence
• Low-dimensional model of turbulent Rayleigh-Benard convection in a Cartesian cell with square domain
• Echo State Network for two-dimensional turbulent moist Rayleigh-Bénard convection
• Reconstructing Rayleigh-Bénard flows out of temperature-only measurements using nudging
• Extreme vorticity events in turbulent Rayleigh-Bénard convection from stereoscopic measurements and reservoir computing

Cited by in corpus (5)

• Reduced-order modeling of two-dimensional turbulent Rayleigh-Bénard flow by hybrid quantum-classical reservoir computing
• Real-time forecasting of chaotic dynamics from sparse data and autoencoders
• Asymmetrically connected reservoir networks learn better
• Data-driven stability analysis in a multi-element supercritical Liquid Oxygen-methane combustor
• Prediction performance of random reservoirs with different topology for nonlinear dynamical systems with different number of degrees of freedom