Review run-time design errors and stability issues, analyze effect of weights on performance, convert unconstrained controller for linear analysis
Once you have created and designed your model predictive controller, you can review it for potential design issues. For more information, see Review Model Predictive Controller for Stability and Robustness Issues.
Review Controller Design
|Compute closed-loop DC gain from output disturbances to measured outputs assuming constraints are inactive at steady state|
|Calculate the value of a performance metric and its sensitivity to the diagonal weights of an MPC controller|
|Size and order of MPC Controller|
|Compute steady-state value of MPC controller plant model state for given inputs and outputs|
- Review Model Predictive Controller for Stability and Robustness Issues
You can detect potential issues with your MPC controller design at the command line and using MPC Designer.
- Test MPC Controller Robustness using MPC Designer
You can test the robustness of your model predictive controller by simulating it with MPC Designer.
- Compute Steady-State Gain
Compute the closed-loop, steady-state gain for each output when a sustained, unit disturbance is added to each output.
- Extract Controller
Obtain a linear state-space model of an unconstrained MPC controller. You can use this model to analyze the frequency response and performance of the controller.
- Compare Multiple Controller Responses Using MPC Designer
You can compare the time-domain and frequency-domain responses of multiple MPC controller designs.
- Adjust Input and Output Weights Based on Sensitivity Analysis
Compute numerical derivatives of a closed-loop cumulated performance index with respect to weights and use them to improve model predictive controller performance.
- Understanding Control Behavior by Examining Optimal Control Sequence
You can analyze the optimal control sequence computed by a model predictive controller at each sample time.