Noise-augmented offline training of ANN unresolved-scale models

An effective way to solve the forced Burgers' equation in a variational multiscale framework is to close the system of equations using unresolved-scale interaction terms predicted by an artificial neural network. The goal of this thesis is to investigate the accuracy and stability of the said system by training a neural network offline with data...

Machine Learning of Atmospheric Turbulence in a Variational Multiscale Model

Today's leading projections of climate change predicate on Atmospheric General Circulation Models (GCMs). Since the atmosphere consists of a staggering range of scales that impact global trends, but computational constraints prevent many of these scales from being directly represented in numerical simulations, GCMs require "parameterisations'' -...