This example shows how to use a spider plot to compare requirement evaluations before and after optimizing the response. You can use a similar procedure to compare the values of sets of design variables.
|On this page…|
Open the Simulink model and load the pre-configured Design Optimization Tool session.
For this example, which uses a distillation column model, the step response requirements are preconfigured and loaded in the model workspace.
Open the distillation model.
sys = 'distillation_demo'; open_system(sys)
Open the Design Optimization Tool.
In the Simulink® model window, select Analysis > Response Optimization.
Alternatively, click the Response Optimization GUI with preloaded data block in the model and skip the next step.
Load the preconfigured Design Optimization Tool session.
Click the Design Optimization tab. In the Open drop-down list, select Open from model workspace. A window opens where you select the Design Optimization Tool session to load. Select distillation_optim and click OK.
The preconfigured step response requirements are loaded in the Design Optimization Tool.
Evaluate the requirement before optimization.
In the Response Optimization tab, click Evaluate Requirements.
A new variable, ReqValues, containing the evaluation of the requirements appears in the Design Optimization Workspace.
When optimizing the model response, you create a set of requirements that it must satisfy. If the requirements are violated, meaning that they evaluate to nonnegative values, the design variables must be optimized. After the optimization, you can compare the original requirement value with the requirement evaluated using the optimized design variable values.
Plot the requirement value before optimization.
In the Data to Plot list, select ReqValues.
In the Add Plot list, select Spider plot.
The plot has an axis for each edge-and-signal combination defined in the distillation_demo/Desired Step Response check block. Points on each axis represent the violation for that signal-edge combination and the plot connects these points to form a closed polygon representing the initial design. Note that some points are negative, representing satisfied constraints, and some positive, representing violated constraints.
Optimize the model.
A new variable, ReqValues1, containing the evaluation of the requirements using the optimized design variables appears in the Design Optimization Workspace.
Compare the requirement values before and after optimization.
In the Data to Plot list, select ReqValues1.
In the Add Plot list, select Spider plot 1.
The optimized requirement values, ReqValues1, are all negative or zero, indicating that all the constraints are satisfied.