Simulink Design Optimization


Simulink Design Optimization

Analyze model sensitivity and tune model parameters

Get Started:

Parameter Estimator

Build accurate plant models by estimating the parameters and states of your Simulink model from test data. Update and tune digital twins of your systems to better represent their current state.

Parameter Estimator App

Interactively import and preprocess your measured data, select model parameters to estimate, perform estimation, and compare and validate estimation results. You can generate MATLAB code from the app to automate the entire process.

Configuration Options

Choose from a variety of derivative-based and global optimization solvers. You can also set parameter ranges, initialize models at steady-state operating points, and accelerate the parameter estimation process using Parallel Computing Toolbox™.

Digital Twin Tuning

Automatically update the parameters of a deployed digital twin model to match the current asset condition. Deploy the parameter estimation workflow using Simulink Compiler™.

Response Optimizer

Optimize model parameters to meet your design requirements and satisfy constraints.

Response Optimizer App

Interactively setup and run optimization problems to tune Simulink model parameters. You can graphically specify multiple design requirements, choose model parameters to optimize, and generate MATLAB code from the app to automate the entire process.

Design Requirements and Constraints

Choose time and frequency-domain requirements such as step-response characteristics, reference signals to track, and Bode magnitude bounds. For frequency-domain requirements the model is linearized using Simulink Control Design. You can also define custom requirements and constraints.

Configuration Options

Improve design robustness by accounting for uncertainty in your model parameters. You can choose optimization solvers, set parameter ranges, initialize models at steady-state operating points, and accelerate the response optimization process using Parallel Computing Toolbox™.

Lookup Tables

Tune lookup tables for applications such as gain-scheduled controllers. You can impose constraints such as monotonicity and smoothness on the lookup table values. Use adaptive lookup tables for solving calibration problems.

Adaptive Lookup Table Using Test Data to Approximate an Engine's Volumetric Efficiency Surface.

Sensitivity Analyzer

Identify which parameters have the greatest impact on your model's behavior. Explore your model's design space to check the robustness of your design and select better initial conditions for parameter estimation and design optimization.

Sensitivity Analyzer App

Interactively create a set of parameter values by sampling probability distributions and perform global sensitivity analysis. Visualize and analyze the results to identify key model parameters. Generate MATLAB code from the app to automate the process.

Sensitivity Analysis and Montecarlo Simulations For Electrical Circuit Model.

Design Space Exploration

Analyze your model’s design space using Montecarlo simulations and design of experiments. This lets you check the robustness of your design, and also determine the impact key model parameters can have on cost functions and design requirements.

Optimization Performance Improvements

Select parameter values that can be good initial conditions for your Parameter Estimator and Response Optimizer app sessions directly from the Sensitivity Analyzer app by visualizing the results of your sensitivity analysis.