2017_USNCCM14 and 2016_CS229
A Spatial Clustering Algorithm for Constructing Local Reduced order Bases for Nonlinear Model Reduction And A Clustering Algorithm for Reduced Order Modeling of Shock Waves
Included is solv_uns.py, which can be used to generate a more unsteady inviscid Burger’s problems. Included is also solv_rom_hrom.py which implements hyper reduction. Included is solv_rom_col_row_sparse_energy_content.py, which runs a ROM with a sparse implementation of both row and column clustering. This setup chooses the number of singular vectors to take using the energy content criteria, while solv_rom_col_row_sparse.py takes a constant number of vectors you've defined. Presented at USNCCM14 in July 2017. First published as a technical report for Stanford CS229
To run the model reduction code, you would need to request the pymortestbed python implementation of reduced order modeling built by Matt Zahr.
A reduced order modeling method for improving online computation time and accuracy using mesh coarsening
The rom_stuff.py files contains the local implementation of a point selection / mesh coarsening method from my AA290 technical report.
More detailed information including reports and presentations at tinarwhite.com