Optimize Model Response

Specify design variables, progress plots and methods, speed up optimization using parallel computing and fast restart, incorporate parameter uncertainty for robustness testing

Simulink^® Design Optimization™ software provides both command-line tools and a graphical Response Optimizer app for optimizing the response of a Simulink model to meet the specified design requirements. You can tune the model parameters to meet multiple design requirements simultaneously. You can also impose custom constraints on parameter values, use parallel computing to speed up optimization, and generate code for deployment.

Apps

Response Optimizer

Optimize model response to satisfy design requirements, test model robustness

Functions

expand all

Configure Cost Function

`sdo.SimulationTest`	Simulation scenario description
`setup`	Configure simulation scenario and model for optimization and evaluation (Since R2024a)
`restore`	Restore simulation scenario and model to their previous state (Since R2024a)
`sdo.setValueInModel`	Set design variable value in model
`sdo.getValueFromModel`	Get design variable value from model

Configure Parameters

`sdo.getParameterFromModel`	Get design variables for optimization
`param.Continuous`	Continuous tunable parameter
`param.Discrete`	Discrete tunable parameter (Since R2022b)

Configure Optimization

`sdo.optimize`	Solve design optimization problem
`sdo.OptimizeOptions`	Optimization option set for `sdo.optimize` function
`sdo.OperatingPointSetup`	Set up steady-state operating point computation
`sdo.getModelDependencies`	List of model file and path dependencies

Topics

Optimization Basics

Design Optimization to Meet Step Response Requirements (GUI)
Optimize controller parameters to meet step response requirements using Response Optimizer.
Design Optimization to Meet Step Response Requirements (Code)
Optimize controller parameters at the command line.
Design Optimization to Track Reference Signal (GUI)
Optimize parameters without adding Signal Constraint blocks to the model.
Discrete-Valued Variables in Response Optimization (GUI)
Use the Response Optimizer app to interactively tune model parameters that are restricted to a discrete, finite set of values.
Discrete-Valued Variables in Response Optimization (Code)
Use response optimization to tune model parameters that are restricted to a discrete, finite set of values.
Design Optimization to Meet Frequency-Domain Requirements (GUI)
This example shows how to tune model parameters to meet frequency-domain requirements using the Response Optimizer app.
Design Optimization to Meet Frequency-Domain Requirements (Code)
This example shows how to tune model parameters to meet frequency-domain requirements, using the sdo.optimize command.
Design Optimization Using Frequency-Domain Check Blocks (GUI)
Optimize model parameters to meet frequency-domain design requirements using the Response Optimizer.
Design Optimization to Meet Time-Domain and Frequency-Domain Requirements (GUI)
Interactively tune a controller to satisfy time-domain and frequency-domain design requirements using the Response Optimizer app.
Write a Cost Function
Write a cost function for parameter estimation, response optimization, or sensitivity analysis. The cost function evaluates your design requirements using design variable values.
Design Optimization Tuning Parameters in Referenced Models (GUI)
This example shows how to tune parameters in referenced models, using the Response Optimizer.
Design Optimization Tuning Parameters in Referenced Models (Code)
This example shows how to tune parameters in referenced models, using the sdo.optimize command.
Response Optimization with Simscape Models
This example shows how to tune the value of a resistor and maximize the power transferred from one part of an electric circuit to another by using the Response Optimizer app.

Steady-State Optimization

Specify Steady-State Operating Point for Response Optimization
An operating point of a dynamic system defines the states and root-level input signals of the model at a specific time.

Custom Objectives

Design Optimization to Meet a Custom Objective (GUI)
This example shows how to optimize a design to meet a custom objective using the Response Optimizer app.
Design Optimization to Meet a Custom Objective (Code)
This example shows how to optimize a design to meet a custom objective using sdo.optimize.
Design Optimization to Meet Custom Signal Requirements (GUI)
Specify a custom requirement on a model signal in the Response Optimizer.

Uncertain Variables

Optimizing Parameters for Robustness
Incorporate the parameter uncertainty to test the robustness of your design.
Specify Custom Signal Objective with Uncertain Variable (GUI)
This example shows how to specify a custom objective function for a model signal.
Design Optimization with Uncertain Variables (Code)
This example shows how to optimize a design when there are uncertain variables.

Speed Up Optimization

Skip Model Simulation Based on Parameter Constraint Violation (GUI)
This example shows how to optimize a design and specify parameter-only constraints that prevent the model from being evaluated in an invalid solution space.
Speed Up Response Optimization Using Parallel Computing
Scenarios when you can speed up optimization using parallel computing, and how the speedup happens.
Use Parallel Computing for Response Optimization
Use parallel computing for response optimization in the app, or at the command line.
Optimizing Time-Domain Response of Simulink Models Using Parallel Computing
This example shows how to use parallel computing to optimize the time-domain response of a Simulink® model.
Use Fast Restart Mode During Response Optimization
This topic shows how to speed up response optimization using Simulink fast restart.
Use Accelerator Mode During Simulations
Simulink Design Optimization software supports Normal and Accelerator simulation modes.
Improving Optimization Performance Using Parallel Computing
This example shows how to improve optimization performance using the Parallel Computing Toolbox™.

Response Optimizer Tasks

Specify Design Variables for Optimization
Specify continuous and discrete variables to tune in the Response Optimizer app, including initial values and allowed ranges or values.
Specify Signals to Log
Specify signals to log in the Response Optimizer.
Create Linearization I/O Sets
Create linearization input/output sets in the Response Optimizer or Sensitivity Analyzer.
Specify Optimization Options
After you configure design variables and design requirements for response optimization in Response Optimizer, you can specify additional options for the optimization process.
Interact with Plots
This topic shows how to interact with plots in the Response Optimizer.
Compare Requirements and Design Variables Using Spider Plot
This example shows how to use a spider plot to compare requirement evaluations before and after optimizing the response.
Save Design Variable Values for Specific Iteration
This example shows how to save the design variable values for specific optimization iterations.
Update Model with Design Variables Set
This example shows how to update a model with a set of design variables.
Save and Load Optimization Sessions
Structure of an optimization session, and saving and loading sessions in the Response Optimizer.

Code Generation

Generate MATLAB Code for Design Optimization Problems (GUI)
This example shows how to automatically generate a MATLAB® function to solve a design optimization problem.

Troubleshooting

Optimization Does Not Make Progress

What to do if the optimization stalls or no changes are seen in parameters values.

Optimization Convergence

What to do if the optimization does not satisfy design requirements or takes a long time to converge near a solution, or if the system response becomes unstable.

Optimization Speed and Parallel Computing

What to do if no speedup is seen with parallel computing, if the results are different, or if the optimization stalls.

Undesirable Parameter Values

What to do if optimization gives undesirable parameter values or violates bounds on values.

Reverting to Initial Parameter Values

How to quit optimizing and revert to original values.

Featured Examples

Design Optimization for Robustness to Model Component Tolerances

Use sensitivity analysis and response optimization and to evaluate how well a model satisfies design requirements and optimize design variables in the presence of uncertainties in model parameters.

Open Live Script

PID Tuning with Actuator Constraints

Use Simulink® Design Optimization™ to tune the gains of the PID controller (Kp, Ki, and Kd) and optimize the step response of the plant. To view the results, use the following steps.

Open Script

LQG Controller Tuning

Use Simulink® Design Optimization™ to optimize the output response of a plant by tuning the LQR gain matrix and feedforward gain. This model includes uncertainty in the plant model and accounts for this uncertainty in the optimization process.

Open Script

Inverted Pendulum Controller Tuning

Use Simulink® Design Optimization™ to optimize the controller of an inverted pendulum. The inverted pendulum is on a cart and the motion of the cart is controlled. The controller's Proportional, Feed-forward and LQR gains are tuned to limit pendulum angle variations and respond to step changes in cart position optimally.

Open Script

Pitch Rate Controller Tuning

Use Simulink® Design Optimization™ to tune the gains of a Digital Pitch Rate Controller and optimize the response of an Aircraft to a step altitude change. The controller includes state derivative and integral feedback. The controller is tuned to satisfy a 10 percent overshoot and 0.9 second rise time step response characteristic.

Open Script

Heat Exchanger Controller Tuning

Use Simulink® Design Optimization™ to optimize the temperature control of a heat exchanger around a temperature set-point.

Open Script

Servomechanism Tuning

Use Simulink® Design Optimization™ to optimize the position controller parameters for a servomotor piston. This model uses blocks from Stateflow®.

Open Script

Phase Lock Loop Tuning

Use Simulink® Design Optimization™ to tune an all-pass filter of a Phase Lock Loop. The filter includes a second-order low pass filter and a feedthrough gain. The feedthrough gain and the second order filter coefficients are tuned to apply a steady-state input to the Voltage Controlled Oscillator (VCO).

Open Script

Stewart Platform Controller Tuning

Use Simulink® Design Optimization™ to optimize the position controller parameters of a Stewart platform. The Stewart platform is modeled using Simscape™ Multibody™ blocks.

Open Script

Surrogate Optimization in Simulink Design Optimization

Use surrogate optimization in Simulink® Design Optimization™ to optimize the design of a hydraulic cylinder.

Open Live Script

Surrogate Optimization Using the Response Optimizer App

Perform design optimization using the surrogate optimization method in the Response Optimizer App.

Open Live Script