Reduced Order Modeling
Reduced order modeling is a technique for reducing the computational complexity or storage requirements of a model while preserving its fidelity within an acceptable range of error. Working with a reduced order model can simplify control design and analysis.
You can create reduced order models (ROMs) of subsystems modeled in Simulink®, including full-order, high-fidelity, third-party simulation models. You can also create ROMs using existing time-domain data.
After collecting the required ROM input/output data, you can train a ROM of a model type available in System Identification Toolbox software such as the nonlinear ARX, Hammerstein-Wiener, and neural state-space (NSS) model types. You can use the ROM you create for system-level desktop simulation, hardware-in-the-loop (HIL) testing, control design, and virtual sensor modeling.
The Reduced Order Modeler app provides a UI workflow for creating ROMs. To use the app, install the Reduced Order Modeler for MATLAB® Support Package by using the instructions in Get and Manage Add-Ons.
Apps
| Reduced Order Modeler | Create reduced order models based on Simulink models, subsystems within models, or simulation data (Since R2025b) |
Topics
Reduced Order Modeling Basics
- Reduced Order Modeling Overview
Reduce computational complexity of models by creating accurate surrogates.
Data-Driven Methods Using UI Workflow
- Reduced Order Model of a Jet Engine Turbine Blade
Create a ROM of a jet engine turbine blade, using the long short-term memory (LSTM) and NSS model types. - Reduced Order Model of an Airframe
Create a ROM of an airframe modeled in Simulink, using the NSS model type. - Reduced Order Modeling of Battery Electric Vehicle Thermal Management System
Create a static ROM of an electric vehicle thermal management system, using the multilayer perceptron (MLP) model type. - Reduced Order Modeling of Subsystems in Engine Model
Create a ROM of the Induction and Combustion subsystems in the Simulink modelenginespeed, using the nonlinear ARX model type. - Reduced Order Model of a Jet Engine Turbine Blade from Data
Create a ROM from data generated by a high-fidelity model, using the NSS model type.
Data-Driven Methods Using Command-Line Workflow
- Nonlinear ARX Model of SI Engine Torque Dynamics
This example describes modeling the nonlinear torque dynamics of a spark-ignition (SI) engine as a nonlinear ARX model. - Hammerstein-Wiener Model of SI Engine Torque Dynamics
This example describes modeling the nonlinear torque dynamics of a spark-ignition (SI) engine as a Hammerstein-Wiener model. - Neural State-Space Model of SI Engine Torque Dynamics
This example describes reduced order modeling (ROM) of the nonlinear torque dynamics of a spark-ignition (SI) engine using a neural state-space model. - Reduced Order Modeling of Electric Vehicle Battery System Using Neural State-Space Model
This example shows a reduced order modeling (ROM) workflow, where you use deep learning to obtain a low-order nonlinear state-space model that serves as a surrogate for a high-fidelity battery model. - Surrogate Modeling Using Gaussian Process-Based NLARX Model
In this example, you replace a hydraulic cavitation cycle model in Simulink with a surrogate nonlinear ARX (NLARX) model to facilitate faster simulation.
Linearization-Based Methods
- Specify Linearization for Model Components Using System Identification (Simulink Control Design)
You can use System Identification Toolbox software to identify a linear system for a model component that does not linearize well, and use the identified system to specify its linearization. - Reduced Order Modeling of a Nonlinear Dynamical System as an Identified Linear Parameter Varying Model
Identify a linear parameter varying reduced order model of a cascade of nonlinear mass-spring-damper systems.
