Main Content

Run Additional Analysis to Reduce Instances of Rational Approximation

This example shows how to reduce the instances of rational approximation by running additional analysis. You analyze a model and during the analysis, Simulink® Design Verifier™ identifies the presence of approximations and the associated objectives are reported as undecided with test case.

You enable the Run additional analysis to reduce instances of rational approximation option to perform additional analysis to confirm the undecided objectives. When you rerun the analysis, Simulink cache that contains the model representation information is reused to perform faster analysis. For more information see Reuse Model Representation for Analysis.

Step 1: Open the Model

The sldvApproximationsExample model results in approximations due to the calculations 1./3 and 2./3 in the Constant block.

Step 2: Perform Test Case Generation Analysis and Review Results

On the Design Verifier tab, click Generate Tests.

After the analysis completes, the Results Summary window displays that one objective is satisfied and one objective is undecided with test case.

To view the detailed analysis report, in the Results Summary window, click HTML. In the report, the Analysis Information chapter lists the approximations that were performed during analysis

The Objective Status chapter gives detailed description of the objectives.

Step 3: Run Additional Analysis by Reusing Cache

The undecided with test case objective is impacted by approximation, and to confirm this objective status you run additional analysis.

(a) On the Design Verifier tab, click Test Generation Settings > Settings.

(b) In the Configurations Parameters dialog box, on the Design Verifier pane, in Advanced parameters, set the Rebuild model representation option to If change is detected and enable Run additional analysis to reduce instances of rational approximation option. Click OK.

Note: If you create a new model, by default, the Rebuild model representation option is set to If change is detected.

(c) To perform test generation analysis, click Generate Tests. The existing cache is validated against the model and the analysis reuses the cache if no change is detected.

The Results Summary window displays that the cached model representation is validated and no change is detected. Hence, the analysis skips the compatibility check and reuses the model representation for analysis.

After the analysis completes, the Results Summary window displays that one objective is satisfied and one objective is unsatisfiable.

Step 4: Review Analysis Results

To view the detailed analysis report, in the Results Summary window, click HTML. In the report, the Objectives Status chapter gives a detailed description of the objectives.

Related Topics