Sparsity-enabled CROM computes cluster-based reduced-order models (CROM) from compressed data and allows one to find few optimized sensor locations tailored to the specific model. Estimating a CROM from those compressed or few point measurements preserves the model structure and topology as compared to model estimated from the full data. The publication is available on arXiv.
- Clone this repository to your desktop.
- Add path to
sparseCROM/srcfolder to Matlab search path using
addpath('<path to mds>/sparseCROM/src').
For determining the optimized sensor locations tailored to a specific CROM, the following packages need to be installed.
Sparse Sensor Placement Optimization (SSPOC), which sets up the optimization problem. It is sufficient to add the file
SSPOC.mto the source folder
The optimization problem is solved using the cvx toolbox, which needs to be installed.
examples/example.m for demonstrating the approach on the period double gyre flow, a simplified model of the gulf stream ocean front. Just execute this file in MatLab and it will generate the plot files in
See the LICENSE file for details.