Pinned Repositories
CMDS-14-impact-resistantance-NN-prediction-
DeepEnergy-TopOpt
Density-based topology optimization via the deep energy method
DEM_for_J2_plasticity
A deep energy method (DEM) to solve J2 elastoplasticity problems in 3D.
Graph_DEM
LatticeOPT
A heuristics-based topology optimization algorithm for thin-walled lattice structures in Abaqus/Explicit
LatticeResponse_NN_Prediction
A recurrent neural network model to predict the stress-strain curve and energy absorption of lattices given its cross section.
ResUNet-DeepONet-Plasticity
Implementation of a ResUNet-based DeepONet for predicting stress distribution on variable input geometries subject to variable loads. A ResUNet is used in the trunk network to encode the variable input geometries, and a feed-forward neural network is used in the branch to encode the loading parameters.
S-DeepONet
A sequential DeepONet model implementation that uses a recurrent neural network (GRU and LSTM) in the branch and a feed-forward neural network in the trunk. The branch network efficiently encodes time-dependent input functions, and the trunk network captures the spatial dependence of the full-field data.
Jasiuk-Research-Group's Repositories
Jasiuk-Research-Group/DEM_for_J2_plasticity
A deep energy method (DEM) to solve J2 elastoplasticity problems in 3D.
Jasiuk-Research-Group/ResUNet-DeepONet-Plasticity
Implementation of a ResUNet-based DeepONet for predicting stress distribution on variable input geometries subject to variable loads. A ResUNet is used in the trunk network to encode the variable input geometries, and a feed-forward neural network is used in the branch to encode the loading parameters.
Jasiuk-Research-Group/S-DeepONet
A sequential DeepONet model implementation that uses a recurrent neural network (GRU and LSTM) in the branch and a feed-forward neural network in the trunk. The branch network efficiently encodes time-dependent input functions, and the trunk network captures the spatial dependence of the full-field data.
Jasiuk-Research-Group/DeepEnergy-TopOpt
Density-based topology optimization via the deep energy method
Jasiuk-Research-Group/LatticeOPT
A heuristics-based topology optimization algorithm for thin-walled lattice structures in Abaqus/Explicit
Jasiuk-Research-Group/Graph_DEM
Jasiuk-Research-Group/LatticeResponse_NN_Prediction
A recurrent neural network model to predict the stress-strain curve and energy absorption of lattices given its cross section.
Jasiuk-Research-Group/Bio-inspired-low-porosity-structures-using-Neural-Networks-GRU-Implemenation
The GRU model, trained to predict stress-strain response and energy absorption, uses eight discrete parameters to characterize the design space. It efficiently predicts new design responses in 0.16 milliseconds, enabling the rapid performance evaluation of 128,000 designs any given strain rate and final strain.
Jasiuk-Research-Group/DeepONet-CrystalPlasticity
Material-Response-Informed DeepONet and its Application to Polycrystal Stress-strain Prediction in Crystal Plasticity
Jasiuk-Research-Group/CMDS-14-impact-resistantance-NN-prediction-
Jasiuk-Research-Group/S-DeepONet-transient-predictions
An improved sequential DeepONet model implementation that uses a recurrent neural network (GRU) in the branch and a feed-forward neural network in the trunk. It can predict the full field solutions at multiple time steps given a time-dependent input function and the domain.