Reduced Order Modeling Support Package (Beta)

Create reduced order models of Simulink models or subsystems, including subsystems that integrate high-fidelity 3rd party tools.

Note that from MATLAB release R2025b Update 3 onward this support package has been replaced by Reduced Order Modeler for MATLAB, use that link to download the most up-to-date and maintained version.

Create reduced order models of Simulink models or subsystems, including subsystems that integrate high-fidelity 3rd party tools. Design experiments to simulate a Simulink model and collect data to train a reduced order model. You can also create ROMs using existing time-domain data. Train nonlinear ARX, neural state-space, LSTM, Multi-Layer Perceptron, and Interpolation models.