Print Email Facebook Twitter Noise-augmented offline training of ANN unresolved-scale models Title Noise-augmented offline training of ANN unresolved-scale models Author Pusuluri, Abhinand (TU Delft Aerospace Engineering) Contributor Hulshoff, Steven (mentor) Degree granting institution Delft University of Technology Programme Aerospace Engineering | Aerodynamics and Wind Energy Date 2020-01-27 Abstract 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 enriched by two different noise -augmentation techniques - White and Gaussian. Subject noise-augmentationWhite noiseGaussian noisevariational multiscale methodArtificial Neural Networks To reference this document use: http://resolver.tudelft.nl/uuid:03b345f6-464a-4dd2-ac50-4fa9c880c839 Part of collection Student theses Document type master thesis Rights © 2020 Abhinand Pusuluri Files PDF A_Pusuluri_MSc_Thesis.pdf 14.92 MB Close viewer /islandora/object/uuid:03b345f6-464a-4dd2-ac50-4fa9c880c839/datastream/OBJ/view