Reservoir computing model of two-dimensional turbulent convection
Sandeep Pandey, Jörg Schumacher
Abstract
Reservoir computing models are one possible architecture of recurrent neural networks. Here, a reservoir computing model is applied to reproduce the low-order statistics of a two-dimensional turbulent Rayleigh-B\'enard flow without solving the underlying Boussinesq equations.
Topics & Concepts
TurbulenceGeologyMechanicsConvectionMeteorologyEnvironmental sciencePhysicsStatistical physicsNeural Networks and Reservoir ComputingModel Reduction and Neural NetworksNeural Networks and Applications